INDEX

Explanations

proper nouns related to people's names

oai_token-act-pair · gpt-3.5-turbo

names of individuals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

rob

-0.69

insula

-0.67

yright

-0.67

ciating

-0.65

 foss

-0.65

vernment

-0.63

IBLE

-0.63

 pencil

-0.62

 Predator

-0.62

 Circle

-0.62

POSITIVE LOGITS

ti

0.95

hak

0.90

onest

0.86

azard

0.85

merga

0.81

mad

0.80

annah

0.80

indu

0.80

yy

0.79

dar

0.78

Activations Density 0.050%

proper nouns related to people's names

names of individuals

No Comments

No Known Activations