INDEX

Explanations

punctuation marks, particularly commas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

ahn

-0.16

angi

-0.15

emoc

-0.15

trinsic

-0.14

orf

-0.14

entin

-0.14

Ĭ¶

-0.14

anto

-0.13

 Snowden

-0.13

POSITIVE LOGITS

utta

0.17

etc

0.17

 ÑĤÐ¾ÑīÐ¾

0.16

 Slee

0.15

spath

0.15

 Halk

0.14

/ay

0.14

reas

0.14

dit

0.13

inae

0.13

Activations Density 0.135%

punctuation marks, particularly commas

No Comments

No Known Activations