INDEX

Explanations

instances of the word "access."

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

access transistor or nodes

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-0.42

 Bohn

-0.42

heartedly

-0.39

the

-0.38

 Publicar

-0.36

 marito

-0.35

oulou

-0.35

 mold

-0.35

iomanip

-0.35

POSITIVE LOGITS

access

1.45

 Access

1.44

Access

1.36

 access

1.36

 ACCESS

1.34

ACCESS

1.33

 accesses

1.09

 accès

1.03

 acceso

1.02

Acces

1.00

Activations Density 0.018%

instances of the word "access."

access transistor or nodes

No Comments

No Known Activations