INDEX

Explanations

verbs indicating actions performed by individuals

oai_token-act-pair · gpt-3.5-turbo

phrases related to accountability and actions performed by individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Entered

-0.82

 suspects

-0.68

 Gork

-0.64

hog

-0.59

tur

-0.58

Det

-0.58

 tongues

-0.57

 tires

-0.56

drawn

-0.55

SIG

-0.55

POSITIVE LOGITS

pez

0.98

 differently

0.83

 wrong

0.78

å¥³

0.76

ournal

0.73

 administr

0.72

etting

0.72

 unconsciously

0.69

hing

0.69

omnia

0.67

Activations Density 0.063%

verbs indicating actions performed by individuals

phrases related to accountability and actions performed by individuals

No Comments

No Known Activations