INDEX

Explanations

words related to the beginning or onset of events or processes

oai_token-act-pair · gpt-3.5-turbo

words indicating the initiation or commencement of actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ighth

-0.77

atform

-0.65

=-=-=-=-=-=-=-=-

-0.64

atu

-0.63

Ago

-0.62

swer

-0.61

 entirety

-0.61

ocally

-0.58

=-=-=-=-

-0.57

berus

-0.55

POSITIVE LOGITS

 behaving

1.04

 noticing

0.96

to

0.96

 disappearing

0.94

 anew

0.93

 creeping

0.88

 appearing

0.88

 leaking

0.88

 piling

0.87

 popping

0.86

Activations Density 0.061%

words related to the beginning or onset of events or processes

words indicating the initiation or commencement of actions or events

No Comments

No Known Activations