INDEX

Explanations

proper nouns related to locations or names of places

oai_token-act-pair · gpt-3.5-turbo

references to a specific location or entity

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

aler

-0.82

naire

-0.77

onde

-0.75

ione

-0.75

orted

-0.75

rand

-0.73

iflower

-0.72

uality

-0.72

inelli

-0.71

andise

-0.70

POSITIVE LOGITS

phies

0.85

cule

0.75

 restraints

0.73

ãĤ´

0.73

 hurdles

0.70

 mosqu

0.69

 slopes

0.69

fet

0.68

ãģĻ

0.68

æĪ¦

0.67

Activations Density 0.070%

proper nouns related to locations or names of places

references to a specific location or entity

No Comments

No Known Activations