INDEX

Explanations

terms related to legal actions or parliamentary processes

oai_token-act-pair · gpt-3.5-turbo

phrases related to legal and political actions or events

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

 horizont

-0.69

 bred

-0.68

 mistake

-0.64

 contradiction

-0.63

 misconception

-0.63

 whatsoever

-0.62

 reproduced

-0.61

 belonged

-0.60

 lied

-0.60

 occurrence

-0.60

POSITIVE LOGITS

0.89

 consolid

0.83

egu

0.79

awaited

0.76

 bids

0.75

 deadline

0.74

0.73

haul

0.72

 prepares

0.71

Activations Density 0.544%

terms related to legal actions or parliamentary processes

phrases related to legal and political actions or events

No Comments

No Known Activations