INDEX

Explanations

references to brand names or product identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

wig

-0.16

urd

-0.15

 noci

-0.14

exus

-0.14

æ¯Ľ

-0.14

å¹¹ç·ļ

-0.14

amik

-0.14

ÑĩÐ¸Ñħ

-0.14

outers

-0.14

POSITIVE LOGITS

 ÑģÑĤÐ°Ð²

0.14

à¸Ńà¸¢

0.14

¢åįķ

0.14

ptest

0.14

FRE

0.14

èĪį

0.14

.hl

0.14

ENU

0.14

oir

0.14

ãĤ·ãĥ§ãĥ³

0.13

Activations Density 0.001%

references to brand names or product identifiers

No Comments

No Known Activations