INDEX

Explanations

references to former roles or statuses of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

aware

-0.17

ian

-0.16

opp

-0.14

mey

-0.14

eling

-0.14

_EXTENSIONS

-0.14

aps

-0.14

¸ı

-0.14

ami

-0.13

Aware

-0.13

POSITIVE LOGITS

/current

0.31

/new

0.20

yme

0.18

/original

0.18

mente

0.18

lies

0.16

lad

0.16

theless

0.15

æ¹¯

0.15

ly

0.15

Activations Density 0.026%

references to former roles or statuses of individuals

No Comments

No Known Activations