INDEX

Explanations

phrases indicating appearances or assumptions about situations or objects

oai_token-act-pair · gpt-3.5-turbo

phrases indicating perceptions or appearances

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

ridge

-0.78

ãĢı

-0.76

ierrez

-0.70

culosis

-0.66

ement

-0.64

tex

-0.64

iamond

-0.63

kefeller

-0.63

arette

-0.63

avorite

-0.62

POSITIVE LOGITS

pires

0.79

piring

0.74

ļé

0.71

ãĥĢ

0.67

XM

0.63

inar

0.61

 insur

0.61

Ĥª

0.59

ENS

0.59

 è£ıç

0.58

Activations Density 0.152%

phrases indicating appearances or assumptions about situations or objects

phrases indicating perceptions or appearances

No Comments

No Known Activations