INDEX

Explanations

instances of apologies and expressions of remorse

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

men

-0.15

la

-0.15

lige

-0.15

Ä±ma

-0.15

lake

-0.14

/types

-0.14

ãĥ¼ãĤ¹

-0.14

ãģĵãĤį

-0.14

iston

-0.14

POSITIVE LOGITS

ylon

0.17

ÑģÐ¾ÑĢ

0.15

apor

0.14

.Factory

0.14

stell

0.14

archy

0.14

uger

0.14

oval

0.14

 prostituer

0.14

nothing

0.13

Activations Density 0.020%

instances of apologies and expressions of remorse

No Comments

No Known Activations