INDEX

Explanations

words related to procedural matters

oai_token-act-pair · gpt-3.5-turbo

terms related to procedures and processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 Bever

-0.75

idas

-0.72

Sales

-0.68

issan

-0.68

asus

-0.66

 Brands

-0.65

Beer

-0.65

 Owner

-0.64

anges

-0.64

Dor

-0.64

POSITIVE LOGITS

 procedural

4.16

 Proced

2.84

 proced

2.64

 jurisd

1.18

 budgetary

1.09

 procedure

1.04

 scripted

1.03

 substantive

1.03

rog

1.00

 procedures

0.96

Activations Density 0.018%

words related to procedural matters

terms related to procedures and processes

No Comments

No Known Activations