INDEX

Explanations

words related to a points system or scoring system

oai_token-act-pair · gpt-3.5-turbo

references to points and scoring systems in a game context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

uthor

-0.91

yss

-0.73

eatures

-0.73

ourt

-0.71

LECT

-0.70

anical

-0.69

ÄŁ

-0.68

 Romeo

-0.68

ther

-0.66

GBT

-0.65

POSITIVE LOGITS

manship

0.98

ense

0.83

points

0.81

holder

0.75

point

0.74

ifications

0.73

 points

0.72

blank

0.71

wagon

0.69

onne

0.69

Activations Density 0.031%

words related to a points system or scoring system

references to points and scoring systems in a game context

No Comments

No Known Activations