INDEX

Explanations

words associated with proper nouns and names

oai_token-act-pair · gpt-3.5-turbo

specific nouns or proper nouns that denote names or titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

VERTISEMENT

-0.46

 Leilan

-0.45

CVE

-0.45

 ãĢĮ

-0.45

)))

-0.44

 SetTextColor

-0.44

 foregoing

-0.44

 mans

-0.44

 recoil

-0.44

 grounds

-0.43

POSITIVE LOGITS

foundland

0.84

tainment

0.71

iversary

0.70

tenance

0.69

assador

0.68

ument

0.68

itage

0.66

usterity

0.66

dinand

0.62

withstanding

0.61

Activations Density 0.672%

words associated with proper nouns and names

specific nouns or proper nouns that denote names or titles

No Comments

No Known Activations