INDEX

Explanations

numeric values related to durations or quantities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

Ã¶h

-0.17

adol

-0.16

iring

-0.15

ÅĻÃŃd

-0.15

Ù«

-0.14

pii

-0.14

privation

-0.14

ceae

-0.14

conti

-0.14

illis

-0.14

POSITIVE LOGITS

 ingress

0.15

/configuration

0.14

upos

0.14

 /*!<

0.14

Î¯Î¿

0.13

 stdClass

0.13

 deficit

0.13

 Wrap

0.13

coat

0.13

Activations Density 0.058%

numeric values related to durations or quantities

No Comments

No Known Activations