INDEX

Explanations

numerical values and time references

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

Tokens before numbers

np_acts-logits-general · gemini-2.0-flash

tokensuka

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

-0.65

in

-0.64

-0.62

for

-0.61

par

-0.61

-0.60

on

-0.60

 over

-0.59

POSITIVE LOGITS

NUMX

1.06

 ་་

1.04

enumi

1.02

 дописавши

1.01

 bezeichneter

0.95

 doubtnut

0.94

tvguidetime

0.93

StoryboardSegue

0.92

}{*}{}

0.91

extAlignment

0.89

Activations Density 0.075%

numerical values and time references

Tokens before numbers

tokensuka

No Comments

No Known Activations