INDEX

Explanations

locations or settings in a narrative, especially those related to criminal activities

oai_token-act-pair · gpt-3.5-turbo

instances of commas in the text

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ocalypse

-0.69

abouts

-0.64

orn

-0.64

 Related

-0.62

rastructure

-0.62

lus

-0.61

 Ratings

-0.60

 Feature

-0.60

soever

-0.59

 Development

-0.57

POSITIVE LOGITS

 intending

1.08

 hoping

1.01

 where

0.95

 which

0.93

 wherein

0.92

 expecting

0.91

 whence

0.91

 whereby

0.90

which

0.82

 retrie

0.81

Activations Density 0.355%

locations or settings in a narrative, especially those related to criminal activities

instances of commas in the text

No Comments

No Known Activations