INDEX

Explanations

phrases that indicate potential actions or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

-0.15

AGER

-0.14

ailer

-0.14

antal

-0.14

 cogn

-0.14

ATORY

-0.14

undred

-0.13

 Redistributions

-0.13

annes

-0.13

mere

-0.13

POSITIVE LOGITS

 having

0.17

ogle

0.16

Ð¾Ð»Ð¾Ñģ

0.15

CAF

0.15

.databinding

0.15

be

0.15

maf

0.14

orch

0.14

 have

0.14

oma

0.14

Activations Density 0.106%

phrases that indicate potential actions or conditions

No Comments

No Known Activations

phrases that indicate potential actions or conditions

No Comments

No Known Activations