INDEX

Explanations

variations and comparisons among different subjects or concepts

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

preceding "that" or "it"

np_acts-logits-general · gemini-2.0-flash

contrasting ideas and exceptions

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

 sitten

-0.54

ános

-0.52

áculos

-0.52

testens

-0.51

だけでも

-0.51

 недавно

-0.48

如果不

-0.48

posedge

-0.48

هرست

-0.47

selbe

-0.47

POSITIVE LOGITS

 unlike

1.17

 Unlike

1.15

unlike

1.12

Unlike

1.11

 Instead

1.04

 instead

1.03

Instead

1.00

不像

0.98

instead

0.95

Rather

0.87

Activations Density 0.439%

variations and comparisons among different subjects or concepts

preceding "that" or "it"

contrasting ideas and exceptions

No Comments

No Known Activations