INDEX

Explanations

occurrences of the year 2015

oai_token-act-pair · gpt-3.5-turbo

references to the year 2015

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 cumbers

-0.77

phia

-0.71

agic

-0.69

oxin

-0.66

ocom

-0.66

osponsors

-0.64

 recip

-0.64

andestine

-0.64

bip

-0.63

ess

-0.63

POSITIVE LOGITS

å¹

0.88

0.76

 onwards

0.74

-'

0.72

âĢ²

0.72

 Patton

0.71

âĶĢâĶĢ

0.71

 Brav

0.70

 Trudeau

0.69

 edition

0.68

Activations Density 0.035%

occurrences of the year 2015

references to the year 2015

No Comments

No Known Activations