INDEX

Explanations

comparative terms indicating degree or intensity

oai_token-act-pair · gpt-3.5-turbo

comparative terms indicating quantities or degrees

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mate

-0.64

 Cust

-0.63

xit

-0.63

 Unknown

-0.60

 Maze

-0.60

se

-0.60

dale

-0.59

Americ

-0.58

Additional

-0.58

uly

-0.57

POSITIVE LOGITS

 than

0.69

omical

0.68

agos

0.66

retch

0.65

":""},{"

0.64

rez

0.62

 grooming

0.61

adesh

0.60

profit

0.59

 qualitative

0.58

Activations Density 0.117%

comparative terms indicating degree or intensity

comparative terms indicating quantities or degrees

No Comments

No Known Activations