INDEX

Explanations

references to keys and key points, especially in contexts involving coding or interaction

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

 itſelf

-1.13

 Theſe

-1.11

 ―――――

-1.11

 Мексичка

-1.10

 themſelves

-1.09

thâu

-1.08

 whoſe

-1.08

 McLaugh

-1.05

 ་་

-1.04

 ſeveral

-1.04

POSITIVE LOGITS

 keys

1.68

 Keys

1.59

key

1.57

KEY

1.49

Key

1.47

KEY

1.41

key

1.40

 KEYS

1.34

keys

1.32

setKey

1.30

Activations Density 0.062%

references to keys and key points, especially in contexts involving coding or interaction

No Comments

No Known Activations