INDEX

Explanations

terms related to typical patterns or standards across various contexts

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

Preceding words of "way," "fashion," or "car."

np_acts-logits-general · gemini-2.0-flash

standard, typical, or usual things

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 AssemblyTitle

-0.86

 للمعارف

-0.72

 CreateTagHelper

-0.65

setcounter

-0.61

 specifically

-0.59

Espèce

-0.58

 Literally

-0.58

 специально

-0.57

 חיצוניים

-0.57

Général

-0.56

POSITIVE LOGITS

 suspects

0.84

nál

0.63

 pattern

0.59

情况下

0.57

endregion

0.56

('*',

0.55

做法

0.54

 fare

0.53

 scenario

0.53

ues

0.53

Activations Density 0.381%

terms related to typical patterns or standards across various contexts

Preceding words of "way," "fashion," or "car."

standard, typical, or usual things

No Comments

No Known Activations