INDEX

Explanations

phrases or words related to causation or consequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

 Sarg

-0.71

LON

-0.70

</b>

-0.69

PVP

-0.68

 Alton

-0.67

 Pree

-0.66

lipop

-0.66

Oma

-0.64

ateral

-0.63

ghed

-0.62

POSITIVE LOGITS

 thus

1.50

 Thus

1.30

thus

1.30

Thus

1.28

 propOrder

1.19

Ainsi

1.19

 hence

1.12

 Hence

1.11

Hence

1.09

 Ainsi

1.09

Activations Density 0.154%

phrases or words related to causation or consequence

No Comments

No Known Activations