INDEX

Explanations

instances of excessive behavior or actions that go too far

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

iec

-0.15

iant

-0.15

allon

-0.14

.bin

-0.14

ëł¹

-0.14

 porno

-0.14

shade

-0.14

ëĭ´

-0.14

ìĭŃìĭľìĺ¤

-0.13

assen

-0.13

POSITIVE LOGITS

zes

0.16

 Mayo

0.14

_AHB

0.14

TOO

0.14

ives

0.14

_TOO

0.14

ãģĬãĤĬ

0.13

lav

0.13

_NEED

0.13

ngine

0.13

Activations Density 0.182%

instances of excessive behavior or actions that go too far

No Comments

No Known Activations