INDEX

Explanations

comparative phrases indicating a lesser amount or degree

oai_token-act-pair · gpt-3.5-turbo

comparative phrases indicating relationships or differences in terms of intensity or degree

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ALE

-0.71

obal

-0.70

bilt

-0.69

erman

-0.66

imity

-0.64

raine

-0.64

ISSION

-0.62

rio

-0.62

 Juda

-0.62

itia

-0.61

POSITIVE LOGITS

 usual

1.39

 anticipated

1.00

usual

1.00

 expected

0.97

 anything

0.93

 ever

0.85

 typical

0.84

 average

0.77

 advertised

0.77

any

0.76

Activations Density 0.089%

comparative phrases indicating a lesser amount or degree

comparative phrases indicating relationships or differences in terms of intensity or degree

No Comments

No Known Activations