INDEX

Explanations

terms related to satisfaction or dissatisfaction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

Data Type

torch.float32

Hook Point

blocks.8.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

 Extras

-0.71

 Sapphire

-0.68

Alz

-0.68

 Birch

-0.67

PLA

-0.66

 overhead

-0.65

 vertically

-0.64

0000000000000000

-0.64

awa

-0.63

 appropriation

-0.62

POSITIVE LOGITS

actory

1.49

ying

1.25

iable

1.12

ied

1.11

atisf

1.09

ished

1.06

iership

1.05

ighed

1.05

liction

1.04

actor

1.04

Activations Density 0.005%

terms related to satisfaction or dissatisfaction

No Comments

No Known Activations