INDEX

Explanations

words related to emitting or emission of different types of energy or substances

oai_token-act-pair · gpt-3.5-turbo

terms related to light and radiation emission

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

anova

-0.92

esan

-0.72

aign

-0.69

apan

-0.69

ivan

-0.68

 Alban

-0.68

udeau

-0.67

essor

-0.67

 Andrews

-0.67

 Chall

-0.65

POSITIVE LOGITS

 emitted

1.08

 emit

1.07

 emitting

1.06

 emission

1.04

 emits

1.02

 emissions

0.94

 emanating

0.87

orescence

0.76

 glow

0.73

getic

0.71

Activations Density 0.023%

words related to emitting or emission of different types of energy or substances

terms related to light and radiation emission

No Comments

No Known Activations