INDEX

Explanations

dates in the format of month/day

oai_token-act-pair · gpt-3.5-turbo

references to the date September 11

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

 Dragonbound

-0.80

 Fletcher

-0.74

 nodd

-0.73

Syl

-0.72

 Slug

-0.67

 Bulg

-0.67

 guid

-0.66

 Sorce

-0.66

 Angelo

-0.66

 Toast

-0.66

POSITIVE LOGITS

oct

1.04

1.03

1.02

1.01

0.99

pmwiki

0.99

0.97

Activations Density 0.037%

dates in the format of month/day

references to the date September 11

No Comments

No Known Activations