INDEX

Explanations

phrases and words related to paving the way for a future event or outcome

oai_token-act-pair · gpt-3.5-turbo

phrases related to initiating or leading to significant changes or developments

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

agara

-0.75

central

-0.63

 Dahl

-0.62

selection

-0.62

inel

-0.61

Leary

-0.61

bite

-0.61

Doctors

-0.60

variable

-0.60

ICES

-0.59

POSITIVE LOGITS

 pave

1.15

 paving

0.95

roleum

0.95

 paved

0.92

\\\\\\\\

0.85

 pavement

0.83

\\\\

0.79

 groundwork

0.78

ments

0.78

lete

0.77

Activations Density 0.014%

phrases and words related to paving the way for a future event or outcome

phrases related to initiating or leading to significant changes or developments

No Comments

No Known Activations