INDEX

Explanations

references to specific products, their features, and user safety information

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

information on policies and services

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

AddTagHelper

-0.65

]")]

-0.57

########.

-0.57

GTCX

-0.55

 nemlig

-0.54

UnusedPrivate

-0.53

ValueStyle

-0.53

 Roskov

-0.52

 zijne

-0.51

exitRule

-0.50

POSITIVE LOGITS

 campaigns

0.49

 policies

0.47

 processes

0.43

 details

0.42

 issues

0.41

 programs

0.40

 schemes

0.39

campaigns

0.39

 fils

0.39

 Campaigns

0.39

Activations Density 0.028%

references to specific products, their features, and user safety information

information on policies and services

No Comments

No Known Activations