INDEX

Explanations

numerical data related to statistics or parameter values

oai_token-act-pair · gpt-3.5-turbo

terms related to statistics

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

¿½

-0.84

 Passage

-0.72

 Atlantis

-0.71

 Clover

-0.69

 Seasons

-0.69

 TRAN

-0.69

Cu

-0.68

 Sierra

-0.66

 Borderlands

-0.65

 Sahara

-0.65

POSITIVE LOGITS

istically

1.16

ements

1.11

uses

1.06

utor

1.05

ics

1.05

uary

1.02

istics

1.02

istical

1.01

ically

0.97

olith

0.97

Activations Density 0.018%

numerical data related to statistics or parameter values

terms related to statistics

No Comments

No Known Activations