INDEX

Explanations

words related to negative connotations or controversies

oai_token-act-pair · gpt-3.5-turbo

words related to deception or harmful intentions

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

schild

-0.75

bley

-0.68

bane

-0.66

Joy

-0.65

ĸļ

-0.62

 beginnings

-0.61

 dove

-0.59

boxing

-0.59

ById

-0.58

enegger

-0.56

POSITIVE LOGITS

ctory

0.98

haps

0.83

cia

0.80

etric

0.77

enza

0.70

cially

0.70

rency

0.69

xia

0.69

iac

0.68

BILITIES

0.67

Activations Density 0.059%

words related to negative connotations or controversies

words related to deception or harmful intentions

No Comments

No Known Activations