INDEX

Explanations

mentions of entities or topics related to news and events

oai_token-act-pair · gpt-3.5-turbo

words related to reporting, documentation, and actions taken

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

 carbohyd

-0.68

 closet

-0.62

 sterile

-0.62

paio

-0.60

 tradem

-0.60

 satell

-0.59

 nostalgic

-0.57

cffff

-0.56

 dismissive

-0.56

 oath

-0.55

POSITIVE LOGITS

 Beir

0.74

lihood

0.70

hower

0.68

Gap

0.67

 Lives

0.66

imon

0.64

 Features

0.62

 Benefits

0.61

 Plays

0.61

 Amendments

0.61

Activations Density 0.520%

mentions of entities or topics related to news and events

words related to reporting, documentation, and actions taken

No Comments

No Known Activations