INDEX

Explanations

ordinal indicators of time or sequence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

emem

-0.17

ierz

-0.17

hlen

-0.16

otron

-0.15

uye

-0.15

otent

-0.14

ummer

-0.14

etal

-0.14

hit

-0.14

ô

-0.14

POSITIVE LOGITS

imals

0.16

NavItem

0.15

avage

0.15

rones

0.14

Pocket

0.14

fur

0.14

ially

0.14

 Teuchos

0.14

wear

0.13

.defaults

0.13

Activations Density 0.010%

ordinal indicators of time or sequence

No Comments

No Known Activations