INDEX

Explanations

quotations within text

oai_token-act-pair · gpt-3.5-turbo

quoted statements or dialogue in the text

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

 metic

-0.72

 derby

-0.71

 killer

-0.70

 developmental

-0.68

 rall

-0.66

 reper

-0.66

 nesting

-0.65

 flared

-0.65

 breakthrough

-0.64

 cancell

-0.64

POSITIVE LOGITS

1.04

there

1.02

Jews

1.02

God

0.99

trust

0.98

straight

0.98

nothing

0.98

Where

0.97

double

0.96

Jewish

0.96

Activations Density 0.066%

quotations within text

quoted statements or dialogue in the text

No Comments

No Known Activations