INDEX

Explanations

expressions of satisfaction or approval towards outcomes and experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 elig

-0.15

onth

-0.14

ìĽĥ

-0.14

 madrid

-0.14

elman

-0.14

 è©ķ

-0.14

Ä¢

-0.14

 Ukra

-0.14

asan

-0.14

POSITIVE LOGITS

 overall

0.18

IPA

0.16

å½¦

0.15

how

0.15

askell

0.14

nd

0.14

extent

0.14

izr

0.14

overall

0.13

 cÃ¡ch

0.13

Activations Density 0.044%

expressions of satisfaction or approval towards outcomes and experiences

No Comments

No Known Activations