INDEX

Explanations

terms related to cognitive abilities and functions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

lo

-0.19

loff

-0.14

old

-0.14

patrick

-0.14

roc

-0.14

ivative

-0.14

wald

-0.14

uder

-0.14

dek

-0.14

ady

-0.14

POSITIVE LOGITS

-be

0.22

 decline

0.19

 reserve

0.17

 impairment

0.17

 Decl

0.16

/memory

0.15

-em

0.15

estr

0.15

rams

0.15

ITIVE

0.15

Activations Density 0.006%

terms related to cognitive abilities and functions

No Comments

No Known Activations