INDEX

Explanations

names "Nicholas"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Nicholas" in various contexts

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

lance

-0.79

vest

-0.78

meet

-0.77

RAY

-0.76

 Gutenberg

-0.76

voc

-0.74

mine

-0.73

naire

-0.70

lez

-0.70

marked

-0.70

POSITIVE LOGITS

olas

1.08

olson

1.04

ols

0.95

ahon

0.84

obar

0.76

ettings

0.75

olls

0.73

raviolet

0.69

 heroism

0.69

fmt

0.69

Activations Density 0.030%

names "Nicholas"

mentions of the name "Nicholas" in various contexts

No Comments

No Known Activations