INDEX

Explanations

specific codes or identifiers

oai_token-act-pair · gpt-3.5-turbo

instances of the word "codes" and its variations in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Inquis

-0.73

ihara

-0.73

 Giant

-0.72

 Fantastic

-0.70

noon

-0.69

ned

-0.67

 Mostly

-0.67

 Flavoring

-0.66

strous

-0.66

tenance

-0.66

POSITIVE LOGITS

 codes

1.49

codes

1.41

 coded

1.14

code

1.12

 Codes

1.08

 code

1.04

oded

1.03

coded

0.95

hare

0.94

otle

0.92

Activations Density 0.009%

specific codes or identifiers

instances of the word "codes" and its variations in various contexts

No Comments

No Known Activations