INDEX

Explanations

mentions of the word "Rat" or words related to treaties and agreements, particularly related to ratification

oai_token-act-pair · gpt-3.5-turbo

terms related to ratification processes

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

ELD

-0.63

 Mandela

-0.63

rity

-0.60

 Hasan

-0.59

IGH

-0.59

 SPACE

-0.59

OUS

-0.58

 wearable

-0.56

conservancy

-0.56

irit

-0.56

POSITIVE LOGITS

chet

1.55

cliffe

1.17

che

1.10

ting

0.98

atos

0.96

ifiers

0.92

chery

0.88

cher

0.87

ios

0.87

ty

0.86

Activations Density 0.017%

mentions of the word "Rat" or words related to treaties and agreements, particularly related to ratification

terms related to ratification processes

No Comments

No Known Activations