INDEX

Explanations

actions or events in a sequence

oai_token-act-pair · gpt-3.5-turbo

phrases indicating actions or processes of transition

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

 Liberties

-0.61

 understatement

-0.60

oubtedly

-0.59

ossier

-0.58

 Doctrine

-0.54

arios

-0.54

framework

-0.54

especially

-0.53

particularly

-0.53

PID

-0.52

POSITIVE LOGITS

 thereafter

0.83

 again

0.81

 abruptly

0.77

again

0.74

 afterwards

0.69

 proceeded

0.69

 mysteriously

0.66

headers

0.64

uddenly

0.63

 afterward

0.62

Activations Density 0.306%

actions or events in a sequence

phrases indicating actions or processes of transition

No Comments

No Known Activations