INDEX

Explanations

discussions about features and improvements in user interfaces or software

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

inia

-0.18

okane

-0.17

avez

-0.15

à¥Ĥà¤®

-0.14

è©

-0.14

anova

-0.14

erte

-0.14

itere

-0.14

ipa

-0.14

aris

-0.13

POSITIVE LOGITS

odont

0.17

 guts

0.16

æ²¿

0.15

 ability

0.15

ouri

0.14

Ear

0.14

 OPTION

0.14

_ele

0.14

 would

0.14

 implement

0.14

Activations Density 0.066%

discussions about features and improvements in user interfaces or software

No Comments

No Known Activations