INDEX

Explanations

numbers related to quantities, specifically the number 2 at a very high activation level

oai_token-act-pair · gpt-3.5-turbo

instances of the number "2" along with its occurrences in varying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

rolet

-0.91

awaru

-0.90

ãĥ¼ãĥĨ

-0.72

etz

-0.71

 Seym

-0.68

mosp

-0.66

efully

-0.65

amaru

-0.65

ians

-0.64

Ik

-0.64

POSITIVE LOGITS

nd

1.66

Â½

1.04

ND

1.01

 thirds

0.94

0.92

0.88

0.87

0.86

0.85

Activations Density 0.091%

numbers related to quantities, specifically the number 2 at a very high activation level

instances of the number "2" along with its occurrences in varying contexts

No Comments

No Known Activations