INDEX

Explanations

mentions of or references to different types of tires

oai_token-act-pair · gpt-3.5-turbo

mentions of tires or related terms

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

adesh

-0.91

rooms

-0.87

ulhu

-0.81

ership

-0.79

ĺħ

-0.73

Ĭ±

-0.72

inel

-0.72

¬¼

-0.71

orld

-0.70

arios

-0.69

POSITIVE LOGITS

 tyre

1.07

 tire

0.98

 tires

0.98

 tread

0.92

 tyres

0.92

 sidew

0.90

senal

0.89

vable

0.88

bike

0.87

anium

0.83

Activations Density 0.009%

mentions of or references to different types of tires

mentions of tires or related terms

No Comments

No Known Activations