INDEX

Explanations

negative phrases or warnings related to behavior and choices

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

AMPLE

-0.15

agy

-0.15

earning

-0.15

anas

-0.15

 rather

-0.14

sein

-0.14

 intra

-0.14

rather

-0.14

ein

-0.14

dh

-0.14

POSITIVE LOGITS

 unless

0.18

 slightest

0.16

ulen

0.16

inaire

0.15

 anymore

0.15

itize

0.15

sed

0.15

nor

0.14

baar

0.14

Ð¾ÑĢÐ°Ð·

0.14

Activations Density 0.253%

negative phrases or warnings related to behavior and choices

No Comments

No Known Activations