INDEX

Explanations

terms related to updating, amending, revising, or removing content within a document or article

oai_token-act-pair · gpt-3.5-turbo

references to administrative or procedural updates and changes within a document

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

etooth

-0.78

 waged

-0.75

 nurture

-0.74

 accelerator

-0.73

 lifestyle

-0.69

ezvous

-0.68

 hospitality

-0.68

icipated

-0.68

stakes

-0.66

 entrepreneurial

-0.65

POSITIVE LOGITS

 edits

1.27

 corrected

1.17

 formatting

1.11

 edit

1.11

 rewritten

1.11

correct

1.10

 revision

1.07

 edited

1.07

 editing

1.06

 originals

1.05

Activations Density 0.893%

terms related to updating, amending, revising, or removing content within a document or article

references to administrative or procedural updates and changes within a document

No Comments

No Known Activations