INDEX

Explanations

terms related to numerical values and specifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

uur

-0.17

noinspection

-0.15

rej

-0.15

sla

-0.15

vou

-0.15

OTP

-0.14

igon

-0.14

serde

-0.14

Latch

-0.14

 Patt

-0.14

POSITIVE LOGITS

 existing

0.17

 Minimum

0.16

minent

0.16

deg

0.16

xed

0.15

ivent

0.15

EMENT

0.14

ford

0.14

ase

0.14

=min

0.14

Activations Density 0.068%

terms related to numerical values and specifications

No Comments

No Known Activations