INDEX

Explanations

references to inserts in various contexts

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

items for insertion

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 Praw

-0.43

Tev

-0.42

Gep

-0.42

ambién

-0.41

Problema

-0.41

 veckor

-0.40

Bew

-0.40

UTERS

-0.40

 electrochemical

-0.38

mgp

-0.38

POSITIVE LOGITS

 insert

1.88

 Insert

1.72

insert

1.70

 inserts

1.63

Insert

1.60

 Inserts

1.46

 inserted

1.38

 INSERT

1.36

inser

1.30

 inserting

1.30

Activations Density 0.007%

references to inserts in various contexts

items for insertion

No Comments

No Known Activations