INDEX

Explanations

words related to a broad range or variety of subjects or entities

oai_token-act-pair · gpt-3.5-turbo

references to the concept of a spectrum across various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hiro

-0.75

verbs

-0.65

BY

-0.64

than

-0.64

ryn

-0.63

earned

-0.62

lisher

-0.62

bear

-0.62

ded

-0.62

Eva

-0.61

POSITIVE LOGITS

 spectrum

1.24

 Spectrum

1.03

geist

0.80

otomy

0.78

atile

0.78

naire

0.78

atility

0.74

 analy

0.73

daq

0.71

 allocations

0.71

Activations Density 0.008%

words related to a broad range or variety of subjects or entities

references to the concept of a spectrum across various contexts

No Comments

No Known Activations