INDEX

Explanations

numeric values and punctuation marks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ÑĩÐ¸ÑĤ

-0.17

inois

-0.14

rdf

-0.14

lld

-0.14

incipal

-0.14

ferred

-0.14

çª

-0.13

à¹ģà¸ľ

-0.13

pired

-0.13

 Pearce

-0.13

POSITIVE LOGITS

hea

0.16

Wunused

0.16

0.15

enze

0.14

>Last

0.14

ÏĥÏĦÎ¬

0.14

æĮĻ

0.14

reetings

0.13

.ibatis

0.13

Activations Density 0.053%

numeric values and punctuation marks

No Comments

No Known Activations