INDEX

Explanations

the word stem "osen" followed by a numerical value

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "rosen" and related variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Experimental

-0.71

 Indie

-0.64

 scrimmage

-0.63

 immersion

-0.62

TM

-0.61

 Dakota

-0.60

 Theft

-0.59

dq

-0.59

spy

-0.59

Ti

-0.58

POSITIVE LOGITS

osen

4.65

olen

2.00

alon

1.28

oren

1.14

isen

1.04

uden

1.03

 Rosen

1.00

oser

0.99

isman

0.97

ocent

0.97

Activations Density 0.023%

the word stem "osen" followed by a numerical value

occurrences of the word "rosen" and related variations

No Comments

No Known Activations