INDEX

Explanations

instances of words related to the action of having or possessing something

oai_token-act-pair · gpt-3.5-turbo

occurrences of the verb "have" in various forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

uti

-0.65

erno

-0.63

oppable

-0.61

iling

-0.60

illation

-0.60

cffffcc

-0.59

fty

-0.59

etting

-0.59

Interested

-0.59

 mathemat

-0.59

POSITIVE LOGITS

 been

1.60

 undergone

1.51

been

1.33

 Been

1.18

 existed

1.15

 survived

1.12

 become

1.07

 disappeared

1.05

 belonged

1.04

 gotten

1.04

Activations Density 0.183%

instances of words related to the action of having or possessing something

occurrences of the verb "have" in various forms

No Comments

No Known Activations