INDEX

Explanations

connectors and understated qualifiers that suggest continuation or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

nge

-0.16

ordes

-0.13

defgroup

-0.13

ilha

-0.13

nj

-0.13

inka

-0.13

bbing

-0.13

erez

-0.13

ÑĭÐ¹

-0.13

fdb

-0.13

POSITIVE LOGITS

far

0.89

far

0.78

Far

0.61

_far

0.59

 date

0.57

Far

0.57

FAR

0.56

-date

0.47

date

0.47

_FAR

0.41

Activations Density 0.043%

connectors and understated qualifiers that suggest continuation or emphasis

No Comments

No Known Activations