INDEX

Explanations

references to Paris and related terms like "parish" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

ovice

-0.18

resse

-0.16

_parsed

-0.15

tors

-0.15

ulses

-0.15

gie

-0.15

gies

-0.15

rak

-0.14

blas

-0.14

_parser

-0.14

POSITIVE LOGITS

arians

0.21

atic

0.19

ners

0.19

ALLEL

0.18

ry

0.17

allax

0.17

ical

0.17

ian

0.16

ly

0.16

adox

0.16

Activations Density 0.050%

references to Paris and related terms like "parish" in various contexts

No Comments

No Known Activations