INDEX

Explanations

terms related to financial evaluations and conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

 shorten

-0.15

jee

-0.15

 Semi

-0.14

 chatt

-0.14

ufen

-0.14

 Ð»Ð¸ÑĪ

-0.14

moy

-0.14

achable

-0.14

éģķ

-0.13

unchecked

-0.13

POSITIVE LOGITS

 minimal

0.44

minimal

0.37

 tiny

0.37

 small

0.36

min

0.35

 limited

0.34

low

0.34

low

0.33

 Minimal

0.31

limited

0.30

Activations Density 0.262%

terms related to financial evaluations and conditions

No Comments

No Known Activations