INDEX

Explanations

words related to specific names, especially "Barron" and "Huntington"

oai_token-act-pair · gpt-3.5-turbo

names of individuals and places within the text

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

 Orioles

-0.79

lihood

-0.78

 Timbers

-0.74

gyn

-0.74

itarian

-0.72

ovic

-0.71

apore

-0.71

 Elys

-0.70

ogenic

-0.69

ication

-0.68

POSITIVE LOGITS

adobe

0.74

berry

0.71

asca

0.70

amily

0.70

ãĥ£

0.70

Mub

0.68

 caul

0.68

 cannabin

0.67

wcs

0.67

ISO

0.67

Activations Density 0.021%

words related to specific names, especially "Barron" and "Huntington"

names of individuals and places within the text

No Comments

No Known Activations