INDEX

Explanations

dialogue or direct speech with quotation marks

oai_token-act-pair · gpt-3.5-turbo

direct quotations or speech within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

avorite

-0.77

 triv

-0.65

 penal

-0.65

 guest

-0.65

 notoriously

-0.64

 weakened

-0.63

 foreground

-0.63

 affected

-0.63

 prone

-0.62

âĹ¼

-0.62

POSITIVE LOGITS

Oh

0.93

Hey

0.92

cow

0.91

Jesus

0.91

hey

0.90

0.89

nothing

0.83

Nothing

0.81

Fuck

0.81

Bring

0.80

Activations Density 0.086%

dialogue or direct speech with quotation marks

direct quotations or speech within the text

No Comments

No Known Activations