INDEX

Explanations

terms related to healthcare and health policy discussions

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs1536-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_1536

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

1,536

Data Type

torch.float32

Hook Point

blocks.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

urat

-0.84

iple

-0.83

raviolet

-0.77

ilts

-0.76

anism

-0.75

-0.74

 Godd

-0.71

steen

-0.71

nered

-0.70

oldemort

-0.69

POSITIVE LOGITS

 industry

1.20

 department

1.14

 ministry

1.14

 sector

1.13

 agency

1.12

 agencies

1.10

 advisory

1.09

 bureau

1.06

cz

1.05

services

1.03

Activations Density 2.530%

terms related to healthcare and health policy discussions

No Comments

No Known Activations