INDEX

Explanations

phrases indicating an explanation or clarification

oai_token-act-pair · gpt-3.5-turbo

phrases that emphasize personal opinion or statement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

"},"

-0.78

icipated

-0.67

\/\/

-0.66

onding

-0.63

berus

-0.63

osure

-0.62

Below

-0.61

shaw

-0.60

illed

-0.59

kefeller

-0.59

POSITIVE LOGITS

 seriously

1.06

 honestly

1.00

 yeah

0.96

 REALLY

0.91

 LOOK

0.86

wow

0.85

yea

0.84

 literally

0.83

htaking

0.78

hey

0.78

Activations Density 0.029%

phrases indicating an explanation or clarification

phrases that emphasize personal opinion or statement

No Comments

No Known Activations