INDEX

Explanations

phrases related to saying goodbye or farewell

oai_token-act-pair · gpt-3.5-turbo

expressions of farewell or goodbye

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

rikes

-0.96

ucket

-0.82

ogun

-0.77

uggest

-0.75

piracy

-0.72

Ranked

-0.72

urg

-0.72

omaly

-0.72

omal

-0.71

awi

-0.70

POSITIVE LOGITS

 goodbye

1.26

bye

1.16

Bye

0.89

 Goodbye

0.85

 farewell

0.83

 hello

0.81

bye

0.80

 greet

0.71

scribe

0.71

 kisses

0.71

Activations Density 0.016%

phrases related to saying goodbye or farewell

expressions of farewell or goodbye

No Comments

No Known Activations