INDEX

Explanations

specific patterns or sequences in scientific data or terminology

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

Letters at the start of words/codes

np_acts-logits-general · gemini-2.0-flash

sequence patterns

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.16.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

 harian

-0.66

MigrationBuilder

-0.62

рои

-0.62

醐

-0.60

 otomatig

-0.59

发表于

-0.59

位

-0.58

ViewFeatures

-0.57

byen

-0.57

ANDUM

-0.56

POSITIVE LOGITS

Azz

0.96

 Coss

0.94

edd

0.88

Yess

0.85

Edd

0.84

Emm

0.82

Utt

0.82

 Genn

0.81

Ayy

0.81

❤️❤️

0.81

Activations Density 0.887%

specific patterns or sequences in scientific data or terminology

Letters at the start of words/codes

sequence patterns

No Comments

No Known Activations