INDEX

Explanations

references to historical figures and locations

oai_token-act-pair · gpt-3.5-turbo

names of historical figures, particularly related to politics and governance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

uality

-0.84

lihood

-0.78

-0.77

thens

-0.71

thening

-0.70

AIN

-0.68

cs

-0.65

urities

-0.63

Spoiler

-0.63

pent

-0.63

POSITIVE LOGITS

 Hoover

1.04

 Truman

0.71

 Randolph

0.70

grim

0.70

acco

0.69

 administration

0.68

ette

0.67

gren

0.67

bum

0.66

stem

0.65

Activations Density 0.006%

references to historical figures and locations

names of historical figures, particularly related to politics and governance

No Comments

No Known Activations