INDEX

Explanations

specific mentions or classifications of entities

oai_token-act-pair · gpt-3.5-turbo

phrases that include the word "as" to indicate classification or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

OTAL

-0.71

reprene

-0.69

rones

-0.69

itiveness

-0.69

SPONSORED

-0.66

apons

-0.66

ITH

-0.65

width

-0.65

astern

-0.65

LET

-0.64

POSITIVE LOGITS

 belonging

1.09

 follows

1.07

 having

0.84

 originating

0.82

 being

0.81

 needing

0.80

 well

0.79

pires

0.76

 favoring

0.74

 possessing

0.74

Activations Density 0.114%

specific mentions or classifications of entities

phrases that include the word "as" to indicate classification or comparison

No Comments

No Known Activations