INDEX

Explanations

expressions of apology or acknowledgment of mistakes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

 stap

-0.15

Aqu

-0.15

aqu

-0.15

iren

-0.14

Ð»Ð¾Ð¼

-0.14

ennie

-0.14

 knob

-0.14

Aqu

-0.13

è¶³

-0.13

ichi

-0.13

POSITIVE LOGITS

 meant

0.21

geme

0.16

 forgot

0.15

CKER

0.15

gend

0.15

EGIN

0.15

 Habit

0.14

tual

0.14

 oversight

0.14

 silly

0.14

Activations Density 0.140%

expressions of apology or acknowledgment of mistakes

No Comments

No Known Activations