INDEX

Explanations

mentions of awards nominations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "nominated" indicating awards or recognitions

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

alities

-0.76

er

-0.74

Assembly

-0.72

port

-0.72

fell

-0.72

ibur

-0.72

 CLSID

-0.71

fare

-0.71

ple

-0.68

hibition

-0.68

POSITIVE LOGITS

 citiz

0.84

ħĭ

0.81

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

0.79

Ń·

0.74

ython

0.72

CLASSIFIED

0.70

ĵĺ

0.68

 pestic

0.67

uled

0.66

ividual

0.65

Activations Density 0.040%

mentions of awards nominations

instances of the word "nominated" indicating awards or recognitions

No Comments

No Known Activations