INDEX

Explanations

phrases related to knowledge, understanding, and awareness

oai_token-act-pair · gpt-3.5-turbo

statements of existence or being

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

itches

-0.70

Footnote

-0.69

 Hitch

-0.66

verts

-0.66

DOI

-0.63

 Moines

-0.62

 Dickinson

-0.61

rites

-0.61

erald

-0.60

 refrain

-0.60

POSITIVE LOGITS

 happening

1.18

 meant

0.92

 supposed

0.90

 going

0.89

 gonna

0.88

nt

0.87

 occurring

0.86

 wrong

0.86

 done

0.84

 presently

0.82

Activations Density 0.090%

phrases related to knowledge, understanding, and awareness

statements of existence or being

No Comments

No Known Activations