INDEX

Explanations

terms related to therapy and thermal processes

oai_token-act-pair · gpt-3.5-turbo

references to therapeutic practices and treatments

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

-0.82

ãĥĥãĥī

-0.79

UTF

-0.68

éĹĺ

-0.66

UAL

-0.65

 Wilde

-0.65

ãģį

-0.65

 Deadpool

-0.64

ITED

-0.64

ocument

-0.64

POSITIVE LOGITS

apist

1.03

 Ther

0.98

apy

0.96

idge

0.93

ron

0.88

igon

0.85

Ther

0.85

imer

0.82

anos

0.81

rha

0.80

Activations Density 0.005%

terms related to therapy and thermal processes

references to therapeutic practices and treatments

No Comments

No Known Activations