INDEX

Explanations

parameters related to dimensions and coordinates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

ajas

-0.16

-scripts

-0.16

vox

-0.15

qua

-0.15

å¡ļ

-0.15

uzzer

-0.15

ÙĤÛĮ

-0.14

resse

-0.14

oku

-0.14

ãĥ³ãĥĦ

-0.14

POSITIVE LOGITS

singleton

0.15

erdale

0.14

ittel

0.14

çĽ®ãĤĴ

0.14

isd

0.14

åŃĺäºİ

0.13

Ð¶Ð´Ð¸

0.13

 Ripple

0.13

.ext

0.13

Activations Density 0.041%

parameters related to dimensions and coordinates

No Comments

No Known Activations