INDEX

Explanations

instances of the word "ladder" in different contexts and scenarios

oai_token-act-pair · gpt-3.5-turbo

mentions of ladders

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

null

-0.94

 Lies

-0.78

 Alic

-0.77

 Extrem

-0.71

asel

-0.70

said

-0.68

 Dragonbound

-0.66

panic

-0.66

fulness

-0.63

 Absent

-0.62

POSITIVE LOGITS

 ladder

1.27

wcs

0.83

pillar

0.83

 crane

0.79

 pige

0.77

 staircase

0.76

ackets

0.76

 craw

0.75

xual

0.75

 stairs

0.75

Activations Density 0.008%

instances of the word "ladder" in different contexts and scenarios

mentions of ladders

No Comments

No Known Activations