INDEX

Explanations

brand names, particularly related to technology products

oai_token-act-pair · gpt-3.5-turbo

occurrences of the abbreviation "AS" and similar terms related to configurations or setups

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

ows

-0.81

naire

-0.79

naires

-0.76

atically

-0.72

ozyg

-0.71

ogue

-0.69

ests

-0.67

oting

-0.66

acies

-0.65

hearted

-0.61

POSITIVE LOGITS

TER

1.07

TERN

0.93

ENA

0.88

daq

0.87

ADA

0.86

ylum

0.85

OAD

0.85

HA

0.84

PAR

0.83

hing

0.82

Activations Density 0.027%

brand names, particularly related to technology products

occurrences of the abbreviation "AS" and similar terms related to configurations or setups

No Comments

No Known Activations