INDEX

Explanations

sentences ending with a specific structure of punctuation

oai_token-act-pair · gpt-3.5-turbo

toxic or harmful statements and concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 Manhattan

-0.74

 Glou

-0.71

Syd

-0.67

 Somerset

-0.67

 scene

-0.64

 Whit

-0.64

Roc

-0.59

 reception

-0.59

 Shattered

-0.59

RAD

-0.58

POSITIVE LOGITS

¬

1.00

âĢł

0.90

agree

0.85

Ĵ

0.85

âĹ¼

0.83

Â¯

0.82

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

0.82

0.80

Â§

0.78

Ãº

0.78

Activations Density 0.230%

sentences ending with a specific structure of punctuation

toxic or harmful statements and concepts

No Comments

No Known Activations