INDEX

Explanations

concepts related to improving performance and evaluation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

culus

-0.16

uve

-0.15

bÃŃ

-0.15

Pun

-0.15

ÐµÑĢÐ²

-0.14

 pseud

-0.13

æµ¦

-0.13

Playlist

-0.13

 tá»ĭch

-0.13

à¥įà¤¯à¤¤

-0.13

POSITIVE LOGITS

 performance

1.05

performance

0.94

 Performance

0.88

 performances

0.88

-performance

0.84

Performance

0.84

 perform

0.82

 PERFORMANCE

0.80

perform

0.80

 Perform

0.76

Activations Density 0.217%

concepts related to improving performance and evaluation in various contexts

No Comments

No Known Activations