INDEX

Explanations

discourse markers that indicate connections or conclusions drawn from earlier statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 Corea

-0.85

 restTemplate

-0.84

 LoginComponent

-0.81

PasswordEncoder

-0.78

THING

-0.76

 Dyck

-0.75

وردار

-0.75

์ตูน

-0.74

 Basili

-0.73

 Phry

-0.73

POSITIVE LOGITS

Hence

1.12

hence

1.00

 hence

0.99

 Hence

0.96

nje

0.75

 Genti

0.72

ddots

0.72

forward

0.71

<h6>

0.70

ENCE

0.69

Activations Density 0.085%

discourse markers that indicate connections or conclusions drawn from earlier statements

No Comments

No Known Activations