INDEX

Explanations

references to internal aspects or functions, often in legal, corporate, or procedural contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ioc

-0.17

iability

-0.16

HeaderValue

-0.16

-shaped

-0.15

ernet

-0.15

Ø§ÙģØª

-0.15

ffect

-0.15

peri

-0.14

ois

-0.14

alyzer

-0.14

POSITIVE LOGITS

/Internal

0.41

/internal

0.35

.Internal

0.25

izado

0.24

ized

0.23

halb

0.22

-facing

0.22

most

0.22

(internal

0.21

izing

0.21

Activations Density 0.028%

references to internal aspects or functions, often in legal, corporate, or procedural contexts

No Comments

No Known Activations