INDEX

Explanations

diverse concepts and qualities associated with novelty and complexity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 ones

-0.19

ÐµÑĤÐµ

-0.16

 hers

-0.16

 mine

-0.16

lit

-0.15

 trop

-0.15

conda

-0.15

fat

-0.15

orta

-0.15

 tire

-0.14

POSITIVE LOGITS

 happens

0.20

 happened

0.19

 happening

0.19

afort

0.17

 authDomain

0.16

rÃ¡l

0.16

 happen

0.16

elow

0.16

 Goldberg

0.15

oling

0.15

Activations Density 0.187%

diverse concepts and qualities associated with novelty and complexity

No Comments

No Known Activations