INDEX

Explanations

statements about dependency and influence 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 @ 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

zk

-0.17

od

-0.14

 Nagar

-0.14

anca

-0.14

rak

-0.14

ife

-0.13

kie

-0.13

aison

-0.13

inet

-0.13

gel

-0.13

POSITIVE LOGITS

requires

0.16

rove

0.15

egan

0.15

asma

0.14

bove

0.14

Peg

0.14

 prem

0.14

çĢ

0.14

PPER

0.14

zeit

0.14

Activations Density 0.173%

statements about dependency and influence in various contexts

No Comments

No Known Activations