INDEX

Explanations

mathematical expressions and definitions related to functions and sequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

theless

-0.20

azzi

-0.18

 Ragnar

-0.15

loff

-0.15

ersen

-0.15

umont

-0.15

ollo

-0.14

ASSES

-0.14

ëĿ½

-0.14

luv

-0.14

POSITIVE LOGITS

 plane

0.16

 integers

0.16

»

0.14

plane

0.14

res

0.14

gs

0.14

 float

0.14

ime

0.14

ais

0.14

egers

0.13

Activations Density 0.052%

mathematical expressions and definitions related to functions and sequences

No Comments

No Known Activations