INDEX

Explanations

numeric values and dates

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

yk

-0.17

ãĥĹãĥª

-0.15

PTS

-0.15

-0.14

rani

-0.14

_COMPLEX

-0.14

Ð°Ð½Ð¾

-0.13

ght

-0.13

mess

-0.13

elles

-0.13

POSITIVE LOGITS

#ad

0.15

immel

0.14

lep

0.14

-UA

0.13

ì¡´

0.13

 occasion

0.13

usan

0.13

 Criterion

0.13

/frontend

0.13

Activations Density 0.009%

numeric values and dates

No Comments

No Known Activations