INDEX

Explanations

mentions of different kinds of therapies in medical or psychological contexts

oai_token-act-pair · gpt-3.5-turbo

references to various forms of therapy

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

abet

-0.80

ership

-0.75

ãĥĥãĥī

-0.73

lihood

-0.70

rake

-0.66

ord

-0.65

clud

-0.64

İ

-0.62

ibli

-0.61

 Eater

-0.61

POSITIVE LOGITS

 sessions

0.94

 therapy

0.89

otherapy

0.85

apist

0.79

utic

0.78

 therapies

0.75

 Therapy

0.75

 regimen

0.74

 session

0.74

ilitating

0.74

Activations Density 0.026%

mentions of different kinds of therapies in medical or psychological contexts

references to various forms of therapy

No Comments

No Known Activations