INDEX

Explanations

past tense verbs ending in 'ed'

oai_token-act-pair · gpt-3.5-turbo

past tense verbs related to recovery or returning

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Bei

-0.81

Nur

-0.80

 Dani

-0.74

nesota

-0.73

 Indra

-0.72

ãĤ¦ãĤ¹

-0.71

Rim

-0.70

 Meadows

-0.69

 Hilbert

-0.69

 Canaver

-0.69

POSITIVE LOGITS

sense

0.80

sov

0.74

paying

0.73

 lawfully

0.70

raid

0.69

 handshake

0.68

 infringing

0.67

happy

0.67

 guiActiveUn

0.67

 orchestr

0.66

Activations Density 0.000%

past tense verbs ending in 'ed'

past tense verbs related to recovery or returning

No Comments

No Known Activations