INDEX

Explanations

the word or concept "later."

oai_token-act-pair · gpt-3.5-turbo

references to the word "later."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

HY

-0.70

hot

-0.70

cking

-0.69

 Cause

-0.68

HO

-0.68

Ble

-0.66

ÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂłÂł

-0.64

washing

-0.64

Ped

-0.64

EO

-0.63

POSITIVE LOGITS

noon

0.87

 satell

0.82

 generations

0.79

 iterations

0.79

 versions

0.78

 mosqu

0.78

osta

0.76

aneously

0.75

 than

0.75

 batches

0.72

Activations Density 0.029%

the word or concept "later."

references to the word "later."

No Comments

No Known Activations