INDEX

Explanations

mentions of the word "Huron"

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Hur" 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 @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

artisan

-0.78

ancial

-0.77

ä»£

-0.74

 Skydragon

-0.74

ynthesis

-0.73

atown

-0.72

nesota

-0.71

elect

-0.71

ħĭ

-0.70

OTE

-0.66

POSITIVE LOGITS

rah

0.97

ving

0.92

ves

0.92

ricane

0.90

ley

0.89

ricanes

0.89

witz

0.87

few

0.86

acan

0.78

ried

0.78

Activations Density 0.010%

mentions of the word "Huron"

mentions of the word "Hur" in various contexts

No Comments

No Known Activations