INDEX

Explanations

words related to legal matters and procedures

oai_token-act-pair · gpt-3.5-turbo

phrases related to approvals, decisions, and regulatory 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

ItemThumbnailImage

-0.87

 thereof

-0.83

.}

-0.82

;;

-0.79

 thereafter

-0.77

âĶĢâĶĢâĶĢâĶĢ

-0.77

Laughs

-0.76

ãĢĤ

-0.76

phasis

-0.76

%;

-0.76

POSITIVE LOGITS

 transgender

1.00

 genetically

0.97

 Donald

0.95

 Ivanka

0.90

 Rohingya

0.82

 Hillary

0.81

 Confederate

0.79

 graphene

0.78

 Melania

0.78

 marijuana

0.77

Activations Density 0.580%

words related to legal matters and procedures

phrases related to approvals, decisions, and regulatory actions

No Comments

No Known Activations