INDEX

Explanations

code-related terms and errors

oai_token-act-pair · gpt-3.5-turbo

terms related to programming or technical keywords

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Blog

-0.58

 COUR

-0.56

 Concepts

-0.55

 Article

-0.54

Jeb

-0.52

 excerpts

-0.52

Web

-0.51

 Editorial

-0.50

 ARTICLE

-0.49

 Webs

-0.49

POSITIVE LOGITS

TPPStreamerBot

0.65

uns

0.64

uggage

0.57

ylum

0.56

otto

0.56

minus

0.56

ĵĺ

0.56

Magikarp

0.55

 mischief

0.54

sson

0.53

Activations Density 0.885%

code-related terms and errors

terms related to programming or technical keywords

No Comments

No Known Activations