INDEX

Explanations

action words associated with investigation and evaluation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ery

-0.47

住了

-0.44

ตรี

-0.42

noy

-0.40

 inan

-0.40

-0.39

OPTS

-0.37

AnimationFrame

-0.37

 rois

-0.37

Cre

-0.37

POSITIVE LOGITS

MLLoader

0.95

 estekak

0.93

 nakalista

0.92

Personendaten

0.89

rawDesc

0.86

 CreateTagHelper

0.80

 EconPapers

0.75

oa̍t

0.74

RectangleBorder

0.74

UserScript

0.73

Activations Density 0.233%

action words associated with investigation and evaluation

No Comments

No Known Activations