INDEX

Explanations

elements related to coding and programming constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

-0.16

chw

-0.15

 tales

-0.15

vable

-0.14

 Barker

-0.14

orne

-0.14

hd

-0.14

arti

-0.13

ies

-0.13

ace

-0.13

POSITIVE LOGITS

afc

0.16

 inval

0.15

argar

0.14

etin

0.14

#af

0.14

ãĥ«ãĤ¯

0.14

 Redistribution

0.14

utton

0.14

/stdc

0.14

Activations Density 0.038%

elements related to coding and programming constructs

No Comments

No Known Activations