INDEX

Explanations

concepts related to belief and the justification of ideas

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

justifying blame excuses

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

UnusedPrivate

-0.39

 Chapin

-0.36

 AppModule

-0.36

ClientRect

-0.35

RequestBuilder

-0.35

 designers

-0.35

 grund

-0.34

zugehen

-0.33

 register

-0.33

 installation

-0.32

POSITIVE LOGITS

 nakalista

0.76

etcode

0.57

Atsauces

0.55

 excuses

0.54

oa̍t

0.52

 betweenstory

0.50

 ujednoznacz

0.49

 հղումներ

0.49

 estekak

0.49

 excuse

0.47

Activations Density 0.038%

concepts related to belief and the justification of ideas

justifying blame excuses

No Comments

No Known Activations