INDEX

Explanations

the word "plete" in various contexts

oai_token-act-pair · gpt-3.5-turbo

variations of the word "complete" and related terms

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

saw

-0.61

 mailing

-0.60

 tele

-0.60

 planning

-0.60

 electricity

-0.59

 eaves

-0.58

 conventions

-0.58

 feelings

-0.57

 INTER

-0.57

 general

-0.57

POSITIVE LOGITS

lete

4.64

letes

3.91

let

1.51

lette

1.44

leted

1.37

lets

1.26

LET

1.10

plete

1.06

lett

1.02

legate

0.98

Activations Density 0.022%

the word "plete" in various contexts

variations of the word "complete" and related terms

No Comments

No Known Activations