INDEX

Explanations

proper nouns or names of places, potentially related to an event or performance

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names and locations

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

eals

-0.70

INAL

-0.68

RTX

-0.68

potion

-0.68

Increases

-0.66

ebook

-0.65

Trend

-0.64

pg

-0.62

ŀ

-0.60

gp

-0.60

POSITIVE LOGITS

atham

1.29

boro

0.97

emouth

0.89

cliffe

0.85

borough

0.82

aston

0.79

mosp

0.76

 Borough

0.76

hurst

0.75

olate

0.73

Activations Density 0.005%

proper nouns or names of places, potentially related to an event or performance

proper nouns, particularly names and locations

No Comments

No Known Activations