INDEX

Explanations

references to final outcomes or completed products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

onga

-0.16

enda

-0.15

 Nack

-0.14

retty

-0.14

 surgeries

-0.14

.yy

-0.14

Ker

-0.14

ker

-0.14

heimer

-0.14

errick

-0.13

POSITIVE LOGITS

à¤¦à¤°

0.18

 outcome

0.15

outcome

0.15

orie

0.15

iore

0.14

antu

0.14

Ã±o

0.14

inspace

0.14

inox

0.14

orph

0.14

Activations Density 0.050%

references to final outcomes or completed products

No Comments

No Known Activations