INDEX

Explanations

conjunctions and coordinating phrases that connect ideas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

æīįèĥ½

-0.15

ains

-0.15

ÐºÐ¾Ð²Ñĸ

-0.13

ocu

-0.13

ICE

-0.13

andom

-0.13

ãĥĮ

-0.13

ugins

-0.13

urus

-0.13

odef

-0.13

POSITIVE LOGITS

nor

0.54

nor

0.43

Nor

0.40

Nor

0.35

 neither

0.35

ä¹Łä¸į

0.28

NOR

0.25

 Neither

0.23

 Ð½Ð¾ÑĢ

0.20

Neither

0.20

Activations Density 0.240%

conjunctions and coordinating phrases that connect ideas

No Comments

No Known Activations