INDEX

Explanations

proper nouns, specifically names like "Joss" and "Graham"

oai_token-act-pair · gpt-3.5-turbo

the word "oss" 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

ãĥ£

-0.79

ufact

-0.71

è¦ļéĨĴ

-0.69

ropolitan

-0.67

mate

-0.65

ric

-0.64

Beast

-0.63

assies

-0.61

RO

-0.61

azon

-0.60

POSITIVE LOGITS

 Whedon

1.23

essed

1.06

essing

1.03

es

0.96

edIn

0.94

enger

0.90

essor

0.90

essors

0.87

ips

0.85

idents

0.83

Activations Density 0.019%

proper nouns, specifically names like "Joss" and "Graham"

the word "oss" in various contexts

No Comments

No Known Activations