INDEX

Explanations

proper nouns related to specific locations and names

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Monte" and associated entities

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

inyl

-0.82

irlf

-0.80

hips

-0.77

inventoryQuantity

-0.74

inning

-0.72

hip

-0.71

Marginal

-0.66

ivities

-0.63

utenberg

-0.62

usher

-0.62

POSITIVE LOGITS

 Carlo

1.09

 Monte

0.92

ucci

0.83

fort

0.83

lette

0.81

lla

0.79

bris

0.79

quist

0.77

jo

0.77

zen

0.77

Activations Density 0.018%

proper nouns related to specific locations and names

mentions of the name "Monte" and associated entities

No Comments

No Known Activations