INDEX

Explanations

words related to the concept of devolution or dissolution

oai_token-act-pair · gpt-3.5-turbo

references to developmental processes or changes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

TPS

-0.73

atton

-0.66

atra

-0.66

 FIGHT

-0.65

utenberg

-0.65

ttp

-0.63

 Jill

-0.62

OUT

-0.62

 Kaufman

-0.62

 Stevens

-0.62

POSITIVE LOGITS

merce

0.86

ciating

0.82

ities

0.76

olved

0.75

fecture

0.73

lly

0.72

ancing

0.72

minist

0.71

ate

0.71

heid

0.70

Activations Density 0.024%

words related to the concept of devolution or dissolution

references to developmental processes or changes

No Comments

No Known Activations