INDEX

Explanations

words related to objects or concepts with significant symbolism

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names of series, locations, and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

etheless

-0.84

anwhile

-0.73

Downloadha

-0.68

gettable

-0.68

 trough

-0.68

arnaev

-0.67

 cleaned

-0.64

nces

-0.62

 bandwagon

-0.61

 ensu

-0.59

POSITIVE LOGITS

amer

0.72

arth

0.71

ihilation

0.70

 Past

0.70

minster

0.68

irc

0.67

areth

0.67

ortality

0.65

equality

0.64

endor

0.64

Activations Density 0.224%

words related to objects or concepts with significant symbolism

proper nouns, particularly names of series, locations, and organizations

No Comments

No Known Activations