INDEX

Explanations

terms related to technology, research, manufacturing, and development

oai_token-act-pair · gpt-3.5-turbo

occurrences of the letter "D" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 seams

-0.68

 balloon

-0.61

ichick

-0.59

 snowball

-0.59

 grape

-0.58

 strengthened

-0.58

 merc

-0.57

 anch

-0.57

holders

-0.56

 entirely

-0.56

POSITIVE LOGITS

3.34

Ds

2.40

DL

1.80

DD

1.80

DM

1.78

1.74

DN

1.73

1.61

DK

1.61

DH

1.61

Activations Density 0.037%

terms related to technology, research, manufacturing, and development

occurrences of the letter "D" in various contexts

No Comments

No Known Activations