INDEX

Explanations

phrases containing the suffix "-im" or "-im" as part of a word

oai_token-act-pair · gpt-3.5-turbo

the occurrences of the substring "im."

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

 Tornado

-0.64

 Norn

-0.64

 outnumbered

-0.63

 roast

-0.62

 Kinnikuman

-0.60

 Oaks

-0.59

 cloves

-0.58

 sten

-0.57

 filler

-0.57

 topped

-0.57

POSITIVE LOGITS

im

4.16

ims

2.26

imb

2.10

IM

2.03

imize

2.02

imi

2.02

imo

1.91

ima

1.83

imov

1.76

imer

1.69

Activations Density 0.017%

phrases containing the suffix "-im" or "-im" as part of a word

the occurrences of the substring "im."

No Comments

No Known Activations