INDEX

Explanations

phrases indicating something or someone being exceptional or a standout in a particular context

oai_token-act-pair · gpt-3.5-turbo

instances of the word "one" with varying contexts indicating unity or singularity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

older

-0.76

ories

-0.70

ourt

-0.63

usra

-0.57

rollers

-0.56

 Lockheed

-0.56

gif

-0.56

inity

-0.56

hawks

-0.56

ãĥ©ãĥ³

-0.55

POSITIVE LOGITS

 sided

0.92

 dimensional

0.81

 Hundred

0.79

 cohesive

0.75

rency

0.75

 hundred

0.75

esan

0.74

dimensional

0.71

Login

0.71

 insepar

0.70

Activations Density 0.076%

phrases indicating something or someone being exceptional or a standout in a particular context

instances of the word "one" with varying contexts indicating unity or singularity

No Comments

No Known Activations