INDEX

Explanations

numerical statistics and comparisons

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

eba

-0.15

 Qualified

-0.14

umu

-0.14

setParameter

-0.13

ondo

-0.13

å¼¥

-0.13

 Sexe

-0.12

ozem

-0.12

oug

-0.12

edes

-0.12

POSITIVE LOGITS

 common

0.52

 normal

0.44

 typical

0.43

common

0.43

-common

0.40

 commonplace

0.40

normal

0.39

.common

0.37

typ

0.37

 usual

0.36

Activations Density 0.195%

numerical statistics and comparisons

No Comments

No Known Activations