INDEX

Explanations

phrases related to criticism or negative evaluation of individuals or situations

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate an additive or cumulative quality in statements

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

DragonMagazine

-1.00

utor

-0.80

neath

-0.74

imester

-0.72

ãĥĥãĥĪ

-0.71

eret

-0.69

imens

-0.69

lux

-0.68

NER

-0.67

DX

-0.67

POSITIVE LOGITS

 vowed

1.45

 urged

1.43

 praised

1.37

 congratulated

1.29

 thanked

1.28

 insisted

1.27

 likened

1.27

 blamed

1.26

 warned

1.26

 reiterated

1.22

Activations Density 0.267%

phrases related to criticism or negative evaluation of individuals or situations

phrases that indicate an additive or cumulative quality in statements

No Comments

No Known Activations