INDEX

Explanations

phrases related to reassignment or reaffirmation

oai_token-act-pair · gpt-3.5-turbo

references to the concept of reassuring or affirming in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

chain

-0.63

 Greeks

-0.63

 accessory

-0.62

bag

-0.61

RAW

-0.61

USD

-0.60

Fi

-0.59

 Filipino

-0.58

gom

-0.58

 conspicuous

-0.57

POSITIVE LOGITS

ining

0.94

igned

0.92

issance

0.92

ension

0.87

essment

0.87

irmed

0.86

ising

0.86

imar

0.85

essed

0.85

ined

0.85

Activations Density 0.046%

phrases related to reassignment or reaffirmation

references to the concept of reassuring or affirming in various contexts

No Comments

No Known Activations