INDEX

Explanations

keywords related to therapy or medical treatment

oai_token-act-pair · gpt-3.5-turbo

terms related to therapeutic treatments and therapies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.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

ULAR

-0.78

 Darkness

-0.76

 Blow

-0.68

 Staten

-0.67

 Deadpool

-0.67

 Forever

-0.66

 Blind

-0.65

 Columbia

-0.65

 Daredevil

-0.64

ELS

-0.63

POSITIVE LOGITS

apist

1.09

apy

1.04

idge

0.98

anos

0.97

rha

0.97

ron

0.96

aton

0.91

icum

0.91

odynamic

0.90

rien

0.88

Activations Density 0.012%

keywords related to therapy or medical treatment

terms related to therapeutic treatments and therapies

No Comments

No Known Activations