INDEX

Explanations

text related to programming and coding, specifically focusing on characters, code snippets, and technical terms

oai_token-act-pair · gpt-3.5-turbo

programming-related terms and syntax

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

enium

-0.79

FML

-0.68

 Olympia

-0.62

imester

-0.61

issance

-0.59

lisher

-0.57

iversal

-0.53

 mainline

-0.52

OPA

-0.52

icip

-0.51

POSITIVE LOGITS

onyms

0.67

 slogans

0.64

igraph

0.60

sed

0.57

 emot

0.56

iae

0.55

 pets

0.55

 accents

0.55

 poems

0.54

 pseudonym

0.54

Activations Density 1.346%

text related to programming and coding, specifically focusing on characters, code snippets, and technical terms

programming-related terms and syntax

No Comments

No Known Activations