INDEX

Explanations

references to hands-on or experiential learning experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

 anún

-0.50

 plainte

-0.48

berdayakan

-0.46

 oignon

-0.44

ementara

-0.44

 seura

-0.44

 cimetière

-0.44

 cejas

-0.44

 celebrid

-0.43

isamment

-0.42

POSITIVE LOGITS

Hands

0.74

 Hands

0.69

hands

0.63

 hands

0.60

 tactile

0.59

 experiential

0.57

IContainer

0.55

 HANDS

0.54

Hand

0.54

 Exper

0.54

Activations Density 0.002%

references to hands-on or experiential learning experiences

No Comments

No Known Activations