INDEX

Explanations

names of individuals, likely related to a specific context or scenario

oai_token-act-pair · gpt-3.5-turbo

specific names or terms associated with various entities, possibly related to locations, people, or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Calder

-0.71

 Clover

-0.68

enegger

-0.63

pim

-0.59

 Reboot

-0.58

 endowed

-0.58

 sterling

-0.57

 silent

-0.55

 imperative

-0.55

BILITY

-0.54

POSITIVE LOGITS

atoon

0.91

ecast

0.87

adic

0.84

pta

0.83

inx

0.80

atchewan

0.78

omore

0.78

oslav

0.78

oku

0.77

iman

0.76

Activations Density 0.124%

names of individuals, likely related to a specific context or scenario

specific names or terms associated with various entities, possibly related to locations, people, or organizations

No Comments

No Known Activations