INDEX

Explanations

phrases indicating prior actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 12-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_12/width_16k/average_l0_82

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 reaſon

-0.77

 pleaſure

-0.77

 ſeveral

-0.77

 Jefus

-0.75

ſelf

-0.74

 purpoſe

-0.70

 raiſ

-0.69

 Conſ

-0.69

ſelves

-0.68

 ſtate

-0.68

POSITIVE LOGITS

 before

0.94

 sebelum

0.94

 πριν

0.83

 bevor

0.82

before

0.82

 før

0.81

BEFORE

0.79

 BEFORE

0.74

 Sebelum

0.74

 voordat

0.74

Activations Density 0.237%

phrases indicating prior actions or events

No Comments

No Known Activations