INDEX

Explanations

phrases related to the correct or appropriate usage of something

oai_token-act-pair · gpt-3.5-turbo

actions related to procedural changes or implementations

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

paio

-0.70

 Vaughan

-0.64

ajo

-0.63

 Trayvon

-0.61

regor

-0.60

eah

-0.59

Book

-0.58

ription

-0.55

whatever

-0.54

ierre

-0.54

POSITIVE LOGITS

 properly

0.94

 individually

0.84

 correctly

0.83

 incorrectly

0.67

 exceeds

0.65

ividually

0.63

unes

0.62

});

0.62

 scientifically

0.62

 prompts

0.61

Activations Density 0.116%

phrases related to the correct or appropriate usage of something

actions related to procedural changes or implementations

No Comments

No Known Activations