INDEX

Explanations

phrases indicating causation or lack of causation between different events or entities

oai_token-act-pair · gpt-3.5-turbo

phrases that discuss the relationship between actions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

worms

-0.76

worm

-0.64

 livest

-0.61

ntil

-0.59

olson

-0.59

shr

-0.57

 peas

-0.55

 helicop

-0.55

 ende

-0.54

 Unle

-0.54

POSITIVE LOGITS

 WITH

0.75

 with

0.74

inson

0.73

ptin

0.68

onies

0.64

 whatsoever

0.62

leeve

0.59

igne

0.59

With

0.59

INK

0.58

Activations Density 0.022%

phrases indicating causation or lack of causation between different events or entities

phrases that discuss the relationship between actions and their implications

No Comments

No Known Activations