INDEX

Explanations

phrases related to chronological events and their sequences

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

following "after" or "before"

np_acts-logits-general · gemini-2.0-flash

facts and sequences

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

-0.74

BeginContext

-0.55

WriteLiteral

-0.52

AnchorStyles

-0.51

+#+

-0.47

[]:

-0.44

WritableDatabase

-0.44

__(/*!

-0.43

 mergeFrom

-0.43

giau

-0.43

POSITIVE LOGITS

 fact

0.99

 hecho

0.84

 fapt

0.79

fact

0.77

 fatto

0.77

 faits

0.75

 فريبيس

0.74

 hechos

0.74

########.

0.69

 fato

0.69

Activations Density 0.150%

phrases related to chronological events and their sequences

following "after" or "before"

facts and sequences

No Comments

No Known Activations