INDEX

Explanations

phrases indicating a search or inquiry

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

kaar

-0.19

eil

-0.17

ÑħÐ¾Ð²

-0.17

gage

-0.17

deaux

-0.16

ksam

-0.16

beck

-0.16

raquo

-0.15

eam

-0.15

.decorate

-0.15

POSITIVE LOGITS

 Encode

0.15

 Shed

0.14

kn

0.14

 knock

0.14

ogo

0.14

loo

0.14

 Lyon

0.13

rec

0.13

Activations Density 0.003%

phrases indicating a search or inquiry

No Comments

No Known Activations