INDEX

Explanations

adjectives related to descriptions and evaluations

oai_token-act-pair · gpt-3.5-turbo

instances of the word "seem" and its conjugations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Exclusive

-0.64

 Pigs

-0.59

 Boxing

-0.56

 Private

-0.55

ance

-0.53

pat

-0.53

Cre

-0.53

 cancer

-0.52

 Final

-0.52

FM

-0.51

POSITIVE LOGITS

 seem

3.27

 appear

2.07

 seems

2.01

 seemed

2.00

 seeming

1.39

 resemble

1.38

 appears

1.33

 Seems

1.27

 appeared

1.09

 tend

1.09

Activations Density 0.012%

adjectives related to descriptions and evaluations

instances of the word "seem" and its conjugations

No Comments

No Known Activations