INDEX

Explanations

the word "Case."

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "Case" in various contexts

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

¥ŀ

-0.95

pez

-0.89

resent

-0.89

emetery

-0.88

 destro

-0.82

srf

-0.81

orough

-0.78

irlf

-0.76

edded

-0.75

avorite

-0.75

POSITIVE LOGITS

cases

1.26

case

1.14

 Case

1.10

Case

0.98

 CASE

0.84

 Cases

0.78

 Keen

0.77

 Files

0.77

 Chimera

0.76

 McCoy

0.75

Activations Density 0.020%

the word "Case."

repetitions of the word "Case" in various contexts

No Comments

No Known Activations