INDEX

Explanations

terms related to the concept of 'flow'

oai_token-act-pair · gpt-3.5-turbo

references to the concept of "flow" 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

 scratched

-0.66

 Telescope

-0.66

andals

-0.65

ises

-0.64

 irrad

-0.62

ifications

-0.62

 kidding

-0.62

 leveled

-0.61

Ü

-0.60

ãĥ¯ãĥ³

-0.60

POSITIVE LOGITS

chart

1.34

 flowing

0.91

flow

0.91

stream

0.90

through

0.88

 flows

0.86

flows

0.86

sheet

0.83

ered

0.83

sheets

0.82

Activations Density 0.044%

terms related to the concept of 'flow'

references to the concept of "flow" in various contexts

No Comments

No Known Activations