INDEX

Explanations

proper nouns or specific names, such as product names, organization names, or movie titles

oai_token-act-pair · gpt-3.5-turbo

references to specific names and titles 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 @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 thereof

-0.87

$.

-0.73

 thereto

-0.72

 respectively

-0.71

 therein

-0.68

}.

-0.59

EStreamFrame

-0.58

 ..."

-0.56

 arrang

-0.55

Interstitial

-0.55

POSITIVE LOGITS

 meanwhile

0.81

 Profile

0.81

 âĵĺ

0.79

 spokesman

0.69

 Edit

0.67

 Emails

0.66

 spokeswoman

0.62

':

0.62

 Releases

0.62

Lam

0.59

Activations Density 1.269%

proper nouns or specific names, such as product names, organization names, or movie titles

references to specific names and titles in various contexts

No Comments

No Known Activations