INDEX

Explanations

terms related to negative evaluations or opinions about issues and situations

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

of interest to

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

MessageTagHelper

-0.47

EDEFAULT

-0.41

oneofs

-0.39

printStats

-0.39

 BufferedWriter

-0.38

ifikationer

-0.38

ErrUnexpectedEOF

-0.37

CastException

-0.37

anmoins

-0.36

fjspx

-0.36

POSITIVE LOGITS

Unknown

0.49

UNKNOWN

0.48

 guste

0.47

itattu

0.47

分から

0.45

httphttps

0.42

Known

0.42

 recognisable

0.41

 intelligible

0.41

 repugnant

0.41

Activations Density 0.077%

terms related to negative evaluations or opinions about issues and situations

of interest to

No Comments

No Known Activations