INDEX

Explanations

time-related actions and durations

oai_token-act-pair · gpt-3.5-turbo

phrases related to personal struggles and resilience

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

anooga

-0.64

were

-0.62

Were

-0.60

 differed

-0.59

 lacked

-0.59

 Ratings

-0.59

didn

-0.57

 Were

-0.57

 were

-0.56

 weren

-0.54

POSITIVE LOGITS

 himself

0.84

 herself

0.80

his

0.69

HIS

0.64

çĶ°

0.62

 tattoos

0.59

 autobiography

0.59

 solo

0.56

 pals

0.54

assi

0.54

Activations Density 1.968%

time-related actions and durations

phrases related to personal struggles and resilience

No Comments

No Known Activations