INDEX

Explanations

punctuation and conjunctive phrases indicating cause and effect relationships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aze

-0.16

arine

-0.15

 Lage

-0.15

otron

-0.15

Ø§Ø²Ùħ

-0.14

å®¿

-0.14

 Mile

-0.13

.cookie

-0.13

AZE

-0.13

 Silk

-0.13

POSITIVE LOGITS

ãĥ³ãĥĩ

0.15

 worse

0.15

eners

0.14

.openg

0.14

threshold

0.14

 lidi

0.14

IBC

0.14

/react

0.14

isay

0.14

reau

0.14

Activations Density 0.245%

punctuation and conjunctive phrases indicating cause and effect relationships

No Comments

No Known Activations