INDEX

Explanations

verbs related to the beginning of an action or process

oai_token-act-pair · gpt-3.5-turbo

the word "begin" and its variations, indicating a repetitive action or initiation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

allo

-0.71

houses

-0.70

rats

-0.67

aths

-0.67

tor

-0.65

ado

-0.65

visor

-0.65

inn

-0.64

outer

-0.64

acho

-0.64

POSITIVE LOGITS

 anew

1.08

 "$:/

0.77

ŃĶ

0.75

mble

0.73

nings

0.73

quire

0.73

0.69

ITIES

0.68

icago

0.68

 ende

0.67

Activations Density 0.044%

verbs related to the beginning of an action or process

the word "begin" and its variations, indicating a repetitive action or initiation

No Comments

No Known Activations