INDEX

Explanations

elements related to self-reflection and personal agency

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

ÑĨÐ¸Ð¾Ð½Ð½ÑĭÐ¹

-0.15

obao

-0.15

iyah

-0.14

dle

-0.14

omite

-0.14

Insensitive

-0.13

ÑĩÐ¸Ð»

-0.13

 ÑĥÑĤÐ¾Ñĩ

-0.13

elib

-0.13

.openConnection

-0.13

POSITIVE LOGITS

 self

0.73

self

0.64

 Self

0.64

-self

0.61

 SELF

0.60

Self

0.58

SELF

0.55

_self

0.54

(self

0.52

	self

0.51

Activations Density 0.296%

elements related to self-reflection and personal agency

No Comments

No Known Activations