INDEX

Explanations

interjections and exclamatory phrases

oai_token-act-pair · gpt-3.5-turbo

use of dialogue or exclamatory actions

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

regate

-0.86

isi

-0.77

irable

-0.77

hare

-0.75

FER

-0.73

ablishment

-0.72

osate

-0.72

Nation

-0.69

hillary

-0.68

redit

-0.68

POSITIVE LOGITS

 grinning

1.21

 clutching

1.21

 blinking

1.16

 staring

1.14

 waving

1.14

 nodding

1.14

 smiling

1.10

 wiping

1.09

 shaking

1.06

 unsure

1.06

Activations Density 0.154%

interjections and exclamatory phrases

use of dialogue or exclamatory actions

No Comments

No Known Activations