INDEX

Explanations

specific programming keywords or commands

oai_token-act-pair · gpt-3.5-turbo

interrogative or introductory phrases indicating conditions or actions

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

CT

-0.61

+/-

-0.61

 guiActiveUnfocused

-0.60

 ï¿½

-0.59

 biod

-0.57

 wheelchair

-0.57

 theirs

-0.56

 float

-0.55

|--

-0.54

 Ecuador

-0.53

POSITIVE LOGITS

resa

1.00

enhagen

0.99

iannopoulos

0.95

ibliography

0.88

mosp

0.85

cloneembedreportprint

0.79

ngth

0.79

odore

0.79

romeda

0.78

llor

0.77

Activations Density 0.528%

specific programming keywords or commands

interrogative or introductory phrases indicating conditions or actions

No Comments

No Known Activations