INDEX

Explanations

phrases related to specific actions or events happening to people in different locations

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "when" within contexts involving actions and events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

ETF

-0.79

ouble

-0.77

Score

-0.77

 hereafter

-0.69

equal

-0.69

ãĥĺ

-0.69

pring

-0.65

Virgin

-0.64

gem

-0.64

-0.63

POSITIVE LOGITS

 suddenly

1.22

 gunfire

1.01

 gunshots

1.00

 sudden

0.97

 gunmen

0.93

 noticed

0.92

 Suddenly

0.90

 tragedy

0.90

uddenly

0.89

 notices

0.88

Activations Density 0.182%

phrases related to specific actions or events happening to people in different locations

occurrences of the word "when" within contexts involving actions and events

No Comments

No Known Activations