INDEX

Explanations

action verbs or phrases indicating a command or directive

oai_token-act-pair · gpt-3.5-turbo

prepositions and directional phrases

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Newsletter

-0.70

DEF

-0.63

Ľ

-0.62

Blue

-0.59

 Finger

-0.59

 Defenders

-0.58

Definition

-0.57

Residents

-0.57

âĿ

-0.57

roller

-0.57

POSITIVE LOGITS

ardless

0.82

arently

0.78

cause

0.75

 whom

0.71

_-

0.70

nown

0.68

ollo

0.66

fortunately

0.66

 Caroline

0.64

 preferably

0.64

Activations Density 0.363%

action verbs or phrases indicating a command or directive

prepositions and directional phrases

No Comments

No Known Activations