INDEX

Explanations

measurements expressed in miles

oai_token-act-pair · gpt-3.5-turbo

references to distances measured in miles

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

sson

-0.93

eers

-0.89

ovich

-0.82

ians

-0.80

 Ducks

-0.79

rification

-0.77

Gaz

-0.76

rition

-0.76

ainment

-0.75

rious

-0.73

POSITIVE LOGITS

\/\/

0.79

 conclud

0.79

$$

0.77

OTH

0.76

mi

0.75

Dee

0.75

aternity

0.75

 compr

0.75

ppel

0.72

OE

0.71

Activations Density 0.024%

measurements expressed in miles

references to distances measured in miles

No Comments

No Known Activations