INDEX

Explanations

identifiers and counts in a data query context

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

eya

-0.15

 localVar

-0.15

JI

-0.14

unate

-0.14

 ç¦

-0.14

kova

-0.14

ÑģÐº

-0.14

oru

-0.14

cest

-0.14

Bound

-0.13

POSITIVE LOGITS

ogram

0.14

åº§

0.14

compact

0.14

Scan

0.14

 Hawkins

0.14

0.13

dre

0.13

pro

0.13

umbing

0.13

ngrx

0.13

Activations Density 0.029%

identifiers and counts in a data query context

No Comments

No Known Activations