INDEX

Explanations

references to "sign" in various contexts, particularly in technical, clinical, or descriptive uses

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

signs and their context

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

 Fletcher

-0.57

期刊论文

-0.55

amba

-0.52

 Herbst

-0.52

perus

-0.52

 Matos

-0.51

McP

-0.51

retto

-0.51

 Fritz

-0.50

-0.49

POSITIVE LOGITS

 Sign

1.05

Sign

1.00

 sign

0.90

SIGN

0.90

sign

0.90

 signo

0.89

 Signs

0.87

 SIGN

0.84

Signs

0.84

 signs

0.78

Activations Density 0.012%

references to "sign" in various contexts, particularly in technical, clinical, or descriptive uses

signs and their context

No Comments

No Known Activations