INDEX

Explanations

terms related to timelines and historical context

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

GEBURTSDATUM

-0.51

Cyfarwyddwr

-0.47

protoimpl

-0.43

 Affiliations

-0.42

GEBURTS

-0.41

 betweenstory

-0.41

ịp

-0.40

offsetof

-0.40

Referências

-0.39

IBAction

-0.39

POSITIVE LOGITS

LikeLike

0.38

habitude

0.36

doit

0.35

 compiti

0.34

IRT

0.33

 appreciation

0.33

fact

0.33

 Arrangements

0.32

Rw

0.32

 interdependence

0.32

Activations Density 0.173%

terms related to timelines and historical context

No Comments

No Known Activations