INDEX

Explanations

terms related to switching and transitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

thâu

-0.85

 труда

-0.80

 שוליים

-0.80

 denaro

-0.79

 selten

-0.79

 uren

-0.77

[].

-0.74

 niemi

-0.74

 صوتيه

-0.74

 humaines

-0.73

POSITIVE LOGITS

 Switch

1.84

 switches

1.81

 switch

1.79

 SWITCH

1.72

Switch

1.68

 Switches

1.68

SWITCH

1.58

 switched

1.56

switches

1.52

switch

1.52

Activations Density 0.103%

terms related to switching and transitions

No Comments

No Known Activations