INDEX

Explanations

words related to categorizing or identifying different types of things

oai_token-act-pair · gpt-3.5-turbo

various types of nouns related to categories or classifications

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

Tik

-0.67

olulu

-0.66

anwhile

-0.64

%);

-0.61

ummies

-0.60

ovember

-0.60

eatures

-0.60

 Reloaded

-0.59

 Shed

-0.59

dq

-0.59

POSITIVE LOGITS

 shenan

0.77

manship

0.75

 arrangement

0.65

ãĤ¬

0.65

Ore

0.64

natureconservancy

0.63

smanship

0.62

Â·Â·

0.62

 guiActiveUnfocused

0.61

yip

0.61

Activations Density 0.298%

words related to categorizing or identifying different types of things

various types of nouns related to categories or classifications

No Comments

No Known Activations