INDEX

Explanations

phrases related to comparisons or choices between different options

oai_token-act-pair · gpt-3.5-turbo

repeated references to the word "the" and its context within phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

maxwell

-0.74

assetsadobe

-0.72

owicz

-0.71

lance

-0.71

govtrack

-0.69

ledge

-0.67

lie

-0.64

Allows

-0.64

ovation

-0.64

REC

-0.63

POSITIVE LOGITS

 aforementioned

1.00

 sexes

0.85

 facets

0.85

 foregoing

0.84

ses

0.81

 factions

0.80

 scenarios

0.80

 options

0.80

 evils

0.79

 extremes

0.79

Activations Density 0.146%

phrases related to comparisons or choices between different options

repeated references to the word "the" and its context within phrases

No Comments

No Known Activations