INDEX

Explanations

phrases or acronyms related to scientific research and development (R&D)

oai_token-act-pair · gpt-3.5-turbo

references to movie ratings and classifications

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

 Vald

-0.73

 Machina

-0.71

 Revel

-0.68

 Leilan

-0.67

ĸļ

-0.65

olulu

-0.65

Kub

-0.65

 Cran

-0.64

ynski

-0.64

 Shak

-0.64

POSITIVE LOGITS

NP

1.18

Gs

1.18

Vs

1.16

DP

1.14

IDs

1.12

Ps

1.10

DF

1.08

NC

1.08

BT

1.08

NF

1.06

Activations Density 0.242%

phrases or acronyms related to scientific research and development (R&D)

references to movie ratings and classifications

No Comments

No Known Activations