INDEX

Explanations

phrases related to human behavior

oai_token-act-pair · gpt-3.5-turbo

mentions of behavior, particularly in a social or psychological context

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

endiary

-0.70

anmar

-0.69

gur

-0.67

enegger

-0.66

vette

-0.66

 Lank

-0.66

ondo

-0.64

mand

-0.64

ixed

-0.64

inite

-0.64

POSITIVE LOGITS

 behaviors

1.00

 behavior

0.98

 modification

0.98

 behaviours

0.96

uation

0.93

 behaviour

0.91

aviour

0.90

avior

0.90

uate

0.89

 patterns

0.89

Activations Density 0.040%

phrases related to human behavior

mentions of behavior, particularly in a social or psychological context

No Comments

No Known Activations