INDEX

Explanations

specific terms related to formal agreements or approvals

oai_token-act-pair · gpt-3.5-turbo

significant nouns related to legal and procedural contexts

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

 guiActiveUn

-0.75

å§«

-0.69

olute

-0.69

axter

-0.68

ayan

-0.65

cipled

-0.65

EStreamFrame

-0.65

ync

-0.64

Ļ½

-0.64

entimes

-0.63

POSITIVE LOGITS

 itself

0.96

 fiasco

0.83

 portion

0.83

 debacle

0.78

osphere

0.75

 brigade

0.74

eers

0.74

 hypothesis

0.72

iest

0.71

 Nebula

0.69

Activations Density 0.492%

specific terms related to formal agreements or approvals

significant nouns related to legal and procedural contexts

No Comments

No Known Activations