INDEX

Explanations

phrases related to self-awareness and comprehension of actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Guill

-0.16

 able

-0.15

inch

-0.14

 Jeffrey

-0.14

 already

-0.14

 notice

-0.14

 sinon

-0.14

plr

-0.14

 coincidence

-0.14

 conducted

-0.13

POSITIVE LOGITS

 signing

0.18

 talking

0.17

 Signing

0.17

 Talking

0.17

 dealing

0.15

ubes

0.15

 Hell

0.15

OTOS

0.14

TintColor

0.14

 Ð·Ð°ÑĤ

0.14

Activations Density 0.040%

phrases related to self-awareness and comprehension of actions

No Comments

No Known Activations

phrases related to self-awareness and comprehension of actions

No Comments

No Known Activations