INDEX

Explanations

terms related to the retail store Wal-Mart

oai_token-act-pair · gpt-3.5-turbo

references to "Wal-Mart" and its various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

assador

-0.75

minus

-0.73

yip

-0.71

ammy

-0.71

resso

-0.71

iler

-0.70

ndra

-0.69

atform

-0.67

ursday

-0.66

idious

-0.65

POSITIVE LOGITS

jong

0.73

wagen

0.68

Raid

0.63

 condos

0.61

Glass

0.60

 Stores

0.59

Loading

0.59

 Downing

0.58

CBC

0.58

pton

0.57

Activations Density 0.310%

terms related to the retail store Wal-Mart

references to "Wal-Mart" and its various contexts

No Comments

No Known Activations