INDEX

Explanations

phrases related to behavioral studies or analysis

oai_token-act-pair · gpt-3.5-turbo

references to behavior 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

ixed

-0.71

zona

-0.69

inka

-0.68

AU

-0.67

 Lank

-0.67

zn

-0.66

endiary

-0.66

inite

-0.66

AZ

-0.66

zynski

-0.65

POSITIVE LOGITS

 modification

1.09

avior

1.05

 behavior

1.00

 behaviors

0.96

uate

0.92

aviour

0.92

uation

0.90

 behavi

0.89

 behaviour

0.87

 behaviours

0.87

Activations Density 0.021%

phrases related to behavioral studies or analysis

references to behavior in various contexts

No Comments

No Known Activations