INDEX

Explanations

code structure or syntax elements

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

 Sachs

-0.16

ÑģÑĤÐµÑĢ

-0.15

let

-0.15

stras

-0.14

èģĬ

-0.14

oulos

-0.14

Bog

-0.13

ÑĦÐµÑĢ

-0.13

hammer

-0.13

semble

-0.13

POSITIVE LOGITS

aint

0.17

yonel

0.15

nesc

0.14

áº¡o

0.14

MapView

0.14

ahn

0.14

eyse

0.13

osl

0.13

ntag

0.13

ono

0.13

Activations Density 0.052%

code structure or syntax elements

No Comments

No Known Activations