INDEX

Explanations

phrases or terms related to criticism or controversy

oai_token-act-pair · gpt-3.5-turbo

instances of the word "offensive" in various contexts

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

chell

-0.83

omed

-0.72

perature

-0.71

Deal

-0.70

ilk

-0.69

Export

-0.68

 Cups

-0.67

aret

-0.67

obook

-0.67

 Cind

-0.66

POSITIVE LOGITS

ity

0.78

ities

0.74

 thrust

0.69

 offensive

0.67

thouse

0.66

 posture

0.63

 bowel

0.63

reb

0.61

ments

0.61

 arte

0.60

Activations Density 0.022%

phrases or terms related to criticism or controversy

instances of the word "offensive" in various contexts

No Comments

No Known Activations