INDEX

Explanations

occurrences of specific non-English characters or symbols

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 Patreon

-0.16

 Brisbane

-0.13

 setup

-0.13

 Karnataka

-0.13

endale

-0.13

CASCADE

-0.13

_CTX

-0.13

 Tasmania

-0.13

umont

-0.13

 Brexit

-0.13

POSITIVE LOGITS

PR

0.35

 Mobil

0.30

PR

0.30

 Herb

0.24

 Corporate

0.23

_PR

0.23

 Public

0.22

CSR

0.22

 corporate

0.22

.PR

0.21

Activations Density 0.003%

occurrences of specific non-English characters or symbols

No Comments

No Known Activations