INDEX

Explanations

phrases referring to justifications or underlying causes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

mailto

-0.60

CppMethod

-0.55

IsPostBack

-0.55

olph

-0.51

rovnik

-0.50

 slutt

-0.49

httphttps

-0.48

 hubiese

-0.47

 vestiti

-0.47

herzog

-0.46

POSITIVE LOGITS

 reason

4.88

reason

4.24

 Reason

4.04

Reason

3.96

 reasons

3.67

 REASON

3.66

reasons

3.25

REASON

3.24

 Reasons

3.19

Reasons

3.03

Activations Density 0.062%

phrases referring to justifications or underlying causes

No Comments

No Known Activations