INDEX

Explanations

phrases that emphasize a particular item or situation

oai_token-act-pair · gpt-3.5-turbo

repetitions of the word "one" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

osponsors

-0.83

inders

-0.82

ories

-0.81

ãĤ¦ãĤ¹

-0.73

allas

-0.73

hips

-0.72

ibility

-0.68

ume

-0.68

ustomed

-0.67

elvet

-0.65

POSITIVE LOGITS

 Hundred

0.88

guy

0.84

 iteration

0.82

 hundred

0.81

 thing

0.76

 eyed

0.74

 liner

0.73

 Thousand

0.70

 sided

0.69

esan

0.69

Activations Density 0.034%

phrases that emphasize a particular item or situation

repetitions of the word "one" in various contexts

No Comments

No Known Activations