INDEX

Explanations

words related to large-scale power consumption or production

oai_token-act-pair · gpt-3.5-turbo

terms related to large data measurements and infrastructure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

enegger

-0.91

 Ambro

-0.80

 Leilan

-0.73

ĸļ

-0.66

QC

-0.65

 omission

-0.65

 Corpus

-0.64

 Dahl

-0.64

Bj

-0.64

 Doct

-0.63

POSITIVE LOGITS

pload

1.05

uries

0.89

illion

0.89

ombat

0.80

rimp

0.78

abytes

0.77

idate

0.74

ãĥ¬

0.73

ixels

0.73

azo

0.73

Activations Density 0.129%

words related to large-scale power consumption or production

terms related to large data measurements and infrastructure

No Comments

No Known Activations