INDEX

Explanations

phrases related to denominating levels of power or importance

oai_token-act-pair · gpt-3.5-turbo

specific terminology related to entertainment industry rankings and awards

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

afety

-0.72

 ACTIONS

-0.69

CTR

-0.67

hops

-0.64

Ĥİ

-0.61

 boundaries

-0.59

gger

-0.59

 advis

-0.58

 Blumenthal

-0.58

 chores

-0.58

POSITIVE LOGITS

ellation

0.77

 Strip

0.73

Led

0.73

ribune

0.72

stru

0.70

liner

0.69

artist

0.69

otos

0.68

 Savior

0.67

Cross

0.66

Activations Density 0.071%

phrases related to denominating levels of power or importance

specific terminology related to entertainment industry rankings and awards

No Comments

No Known Activations