INDEX

Explanations

references to significant legal or political events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

utenberg

-0.20

 Gutenberg

-0.19

Pixels

-0.14

 zorun

-0.14

itchens

-0.14

Ð¾Ð»Ð¾

-0.14

sampling

-0.14

 caffe

-0.14

 Pixels

-0.14

.sax

-0.14

POSITIVE LOGITS

 pardon

0.45

 pard

0.43

 commute

0.36

 commuting

0.36

 comm

0.33

pard

0.33

 Clem

0.31

 granting

0.26

len

0.26

 forgiveness

0.26

Activations Density 0.066%

references to significant legal or political events

No Comments

No Known Activations