INDEX

Explanations

names or terms related to geographical locations

oai_token-act-pair · gpt-3.5-turbo

patterns or sequences of letters, particularly "gu" and similar combinations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

microsoft

-0.68

noon

-0.65

lass

-0.64

growth

-0.62

staking

-0.62

hemat

-0.59

 Rivals

-0.58

 Motorsport

-0.57

 finance

-0.57

pool

-0.56

POSITIVE LOGITS

pta

1.21

llah

0.92

irus

0.92

pload

0.91

irrel

0.87

Ã±a

0.85

atl

0.84

isu

0.82

issance

0.81

illac

0.80

Activations Density 0.096%

names or terms related to geographical locations

patterns or sequences of letters, particularly "gu" and similar combinations

No Comments

No Known Activations