INDEX

Explanations

terms that indicate positive attributes or assessments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

GLOBALS

-0.17

 principle

-0.15

 Principle

-0.15

 somehow

-0.14

ritten

-0.14

ongan

-0.13

oure

-0.13

ville

-0.13

baugh

-0.13

 Leigh

-0.13

POSITIVE LOGITS

æĳ

0.14

rcode

0.14

Ð²Ð¸Ñĩ

0.14

 Trem

0.14

yles

0.14

_exports

0.14

usters

0.14

 éĸ¢éĢ£

0.14

/thumb

0.14

èĴ

0.13

Activations Density 0.076%

terms that indicate positive attributes or assessments

No Comments

No Known Activations