INDEX

Explanations

references to specific entities or organizations marked with 'G'

oai_token-act-pair · gpt-3.5-turbo

references to the letter "G" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

 craw

-0.67

htt

-0.66

terday

-0.63

 sight

-0.63

oÄŁ

-0.62

nsics

-0.62

ĸļ

-0.61

 derail

-0.59

 Razor

-0.59

 retrospect

-0.59

POSITIVE LOGITS

reet

1.20

ossip

1.18

ourmet

1.17

reetings

1.17

ospels

1.15

irlfriend

1.14

oliath

1.14

asp

1.13

opher

1.12

eeks

1.11

Activations Density 0.047%

references to specific entities or organizations marked with 'G'

references to the letter "G" in various contexts

No Comments

No Known Activations