INDEX

Explanations

conditional phrases that suggest expectations or advice

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

obec

-0.16

 flame

-0.14

Ware

-0.14

 Peters

-0.14

onian

-0.14

 paraph

-0.13

ask

-0.13

 behind

-0.13

ãģıãĤĮ

-0.13

BUR

-0.13

POSITIVE LOGITS

atile

0.17

ilogy

0.16

specs

0.15

ourt

0.15

amage

0.15

çĶŁãģį

0.15

ordova

0.15

etak

0.14

alama

0.14

ilot

0.14

Activations Density 0.017%

conditional phrases that suggest expectations or advice

No Comments

No Known Activations