INDEX

Explanations

phrases and words related to medical conditions and treatments

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_fs3072-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_3072

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

3,072

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

OUGH

-0.76

pper

-0.74

etry

-0.71

otide

-0.70

azaki

-0.69

esan

-0.68

elf

-0.68

AMI

-0.67

arium

-0.67

obyl

-0.66

POSITIVE LOGITS

ãĤ¿

0.87

 volcano

0.84

ment

0.78

 participation

0.75

 newsp

0.74

activity

0.74

imental

0.71

ãĤ±

0.71

#$

0.70

iences

0.70

Activations Density 0.975%

phrases and words related to medical conditions and treatments

No Comments

No Known Activations