INDEX

Explanations

terms related to conditions and classifications that imply a level of achievement or quality

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

 Artifact

-0.15

 verses

-0.15

zik

-0.15

 Eaton

-0.15

 Chap

-0.15

Ł³

-0.14

 Briggs

-0.14

 Jonas

-0.13

rike

-0.13

zet

-0.13

POSITIVE LOGITS

0.14

rieved

0.14

/lic

0.14

 komb

0.13

Oz

0.13

omin

0.13

EH

0.13

ochond

0.13

Sud

0.13

èĳ

0.13

Activations Density 0.032%

terms related to conditions and classifications that imply a level of achievement or quality

No Comments

No Known Activations