INDEX

Explanations

terms related to content consumption and preferences

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

 tisk

-0.16

anka

-0.15

anza

-0.14

ofilm

-0.14

å¤§åĪ©

-0.14

rud

-0.13

TEX

-0.13

 Silver

-0.13

 premature

-0.13

POSITIVE LOGITS

ACHER

0.16

acher

0.15

/cms

0.14

 Sour

0.14

ÑĨÐ¸Ð½

0.14

 jose

0.14

ype

0.14

umes

0.14

joy

0.13

odash

0.13

Activations Density 0.216%

terms related to content consumption and preferences

No Comments

No Known Activations