INDEX

Explanations

names or terms related to a specific entity named "Dur"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Dur" 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

NetMessage

-0.97

ħĭ

-0.83

 Simulator

-0.76

 Helpful

-0.69

 Panic

-0.69

pocket

-0.68

insula

-0.68

-0.66

ocene

-0.64

 Ethics

-0.63

POSITIVE LOGITS

Dur

0.97

locks

0.92

rell

0.92

ably

0.87

abl

0.87

anty

0.87

lock

0.87

rier

0.86

uti

0.86

nit

0.85

Activations Density 0.016%

names or terms related to a specific entity named "Dur"

mentions of the name "Dur" in various contexts

No Comments

No Known Activations