INDEX

Explanations

proper nouns, specifically focusing on names such as Molly, Crowley, Zelnick, and OPEC

oai_token-act-pair · gpt-3.5-turbo

names of people and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

ributed

-0.83

 nomine

-0.81

ribution

-0.79

ovation

-0.77

inates

-0.73

iour

-0.73

essee

-0.71

ection

-0.65

ribut

-0.64

ribute

-0.63

POSITIVE LOGITS

zona

0.75

glers

0.73

azine

0.72

wake

0.72

thia

0.71

 Morg

0.71

dule

0.70

eker

0.68

hran

0.68

ãĥĥãĥī

0.68

Activations Density 0.023%

proper nouns, specifically focusing on names such as Molly, Crowley, Zelnick, and OPEC

names of people and organizations

No Comments

No Known Activations