INDEX

Explanations

questions related to individual identity and contributions within a community

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

buah

-0.15

cookie

-0.14

printStats

-0.14

klu

-0.14

hart

-0.14

urry

-0.14

éº

-0.13

lew

-0.13

 delt

-0.13

.throw

-0.13

POSITIVE LOGITS

edd

0.14

ous

0.14

ilities

0.14

icion

0.13

heel

0.13

 Dank

0.13

Na

0.13

hend

0.13

Na

0.13

Row

0.13

Activations Density 0.062%

questions related to individual identity and contributions within a community

No Comments

No Known Activations