INDEX

Explanations

phrases that express nuances in arguments or discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

Âł

-0.15

plain

-0.14

-keys

-0.14

itches

-0.14

lic

-0.14

vailable

-0.14

amburg

-0.14

INLINE

-0.14

-0.13

POSITIVE LOGITS

eka

0.17

ãĥ¬ãĥ³

0.15

aal

0.14

 ÐºÐ¾Ð²

0.14

ãģĭãģĹ

0.14

uma

0.14

elter

0.14

Î¬Î½Î¹

0.14

 Conditioning

0.14

 CONDITION

0.13

Activations Density 0.114%

phrases that express nuances in arguments or discussions

No Comments

No Known Activations