INDEX

Explanations

negative or contrasting terms related to data or comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

errat

-0.17

-0.15

oir

-0.15

itÃ¤

-0.15

-0.14

esan

-0.14

ippi

-0.14

uru

-0.14

 pulse

-0.13

POSITIVE LOGITS

agos

0.14

//=

0.14

oub

0.14

estone

0.14

.newBuilder

0.14

 inval

0.14

finity

0.14

.HorizontalAlignment

0.13

gether

0.13

Ð¼ÐµÐ½Ð°

0.13

Activations Density 0.037%

negative or contrasting terms related to data or comparisons

No Comments

No Known Activations