INDEX

Explanations

phrases related to expert panels or discussions

oai_token-act-pair · gpt-3.5-turbo

references to panels and panelists involved in discussions or expert groups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

dx

-0.74

DIT

-0.69

 Pyro

-0.65

yll

-0.65

 Chains

-0.62

Pip

-0.62

velt

-0.61

ensor

-0.60

 Campaign

-0.60

 Dame

-0.60

POSITIVE LOGITS

ists

1.12

chairs

1.03

igans

0.94

 panels

0.86

ist

0.85

members

0.82

 chairs

0.80

isode

0.80

 convened

0.80

 panel

0.80

Activations Density 0.014%

phrases related to expert panels or discussions

references to panels and panelists involved in discussions or expert groups

No Comments

No Known Activations