INDEX

Explanations

keywords related to enabling and disabling features or modes

oai_token-act-pair · gpt-3.5-turbo

words related to enabling and disabling features or modes in a technical context

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

apest

-1.13

fare

-0.91

ivan

-0.91

von

-0.78

Ã¡n

-0.77

hran

-0.76

dor

-0.76

oled

-0.76

arnaev

-0.75

esan

-0.74

POSITIVE LOGITS

 defaults

0.90

 automatic

0.90

 javascript

0.82

 default

0.78

Disable

0.77

 autop

0.76

 permissions

0.75

 override

0.74

 tresp

0.74

 blinking

0.73

Activations Density 0.034%

keywords related to enabling and disabling features or modes

words related to enabling and disabling features or modes in a technical context

No Comments

No Known Activations