INDEX

Explanations

phrases related to methods, tools, or mechanisms

oai_token-act-pair · gpt-3.5-turbo

phrases that denote methods or means to achieve certain ends or actions

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

oos

-0.71

ONSORED

-0.71

endas

-0.66

aples

-0.65

halla

-0.63

@#&

-0.63

 corrid

-0.63

 suspic

-0.63

YL

-0.62

ADRA

-0.61

POSITIVE LOGITS

terday

0.84

 whereby

0.78

finding

0.70

chery

0.69

forward

0.68

fare

0.68

ivism

0.67

icles

0.67

ãĥŁ

0.66

mith

0.65

Activations Density 0.028%

phrases related to methods, tools, or mechanisms

phrases that denote methods or means to achieve certain ends or actions

No Comments

No Known Activations