INDEX

Explanations

research studies and clinical reports

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

hoeddwyd

-0.64

 queſta

-0.56

 houſe

-0.56

parsedMessage

-0.55

Personensuche

-0.55

Tikang

-0.55

 ExecuteAsync

-0.55

########.

-0.54

rrggbb

-0.54

 createSlice

-0.52

POSITIVE LOGITS

Rev

0.45

 Case

0.40

schlag

0.40

 experience

0.40

Mel

0.39

GraphicsUnit

0.39

appcompat

0.39

✎

0.39

 technique

0.38

 technical

0.38

Activations Density 0.030%

research studies and clinical reports

No Comments

No Known Activations