INDEX

Explanations

constructs related to programming or data manipulation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

udden

-0.16

obra

-0.16

 Ting

-0.14

ials

-0.13

adin

-0.13

Anc

-0.13

 motivation

-0.13

 pert

-0.13

far

-0.13

udence

-0.13

POSITIVE LOGITS

iquer

0.18

mez

0.15

bars

0.15

.Dto

0.15

aisy

0.15

 Leaf

0.15

ambre

0.15

Mesh

0.14

azer

0.14

.firebaseio

0.14

Activations Density 0.119%

constructs related to programming or data manipulation

No Comments

No Known Activations