INDEX

Explanations

tools or instruments

oai_token-act-pair · gpt-3.5-turbo

references to various tools and utilities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 Sons

-0.77

ometown

-0.75

orus

-0.69

uates

-0.68

oning

-0.66

erest

-0.64

nen

-0.64

noon

-0.63

owship

-0.63

liction

-0.62

POSITIVE LOGITS

tips

1.20

kit

1.20

 tools

1.19

 tool

0.96

 guiActiveUn

0.94

*/(

0.87

 Tools

0.86

tools

0.83

 levers

0.81

tip

0.81

Activations Density 0.010%

tools or instruments

references to various tools and utilities

No Comments

No Known Activations