INDEX

Explanations

names or proper nouns

oai_token-act-pair · gpt-3.5-turbo

proper names, particularly those of individuals and locations

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

GY

-0.68

gest

-0.67

Bab

-0.67

 Judah

-0.67

====

-0.67

glers

-0.62

 bowel

-0.62

BIT

-0.62

abil

-0.61

cean

-0.61

POSITIVE LOGITS

intendent

0.94

arnaev

0.86

gomery

0.84

ignty

0.80

anchester

0.78

theless

0.78

ivas

0.78

iard

0.77

ipeg

0.76

ablishment

0.75

Activations Density 0.066%

names or proper nouns

proper names, particularly those of individuals and locations

No Comments

No Known Activations