INDEX

Explanations

references to the origin or cause of something

oai_token-act-pair · gpt-3.5-turbo

terms related to information sources or origins

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Ħ¢

-0.95

interstitial

-0.91

 satell

-0.90

aeper

-0.88

uckle

-0.77

 lapt

-0.76

psey

-0.71

ĵĺ

-0.69

 councill

-0.68

©¶æ¥µ

-0.67

POSITIVE LOGITS

 source

1.26

source

1.23

 sources

1.19

 Sources

1.10

Sources

0.92

Fed

0.90

ources

0.90

 sourcing

0.85

 Source

0.84

forge

0.80

Activations Density 0.018%

references to the origin or cause of something

terms related to information sources or origins

No Comments

No Known Activations