INDEX

Explanations

links or locations within a text

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the availability or location of information or items

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

tons

-0.87

 boundaries

-0.66

 typh

-0.64

 correctness

-0.64

 antidepressants

-0.63

uked

-0.62

 judgment

-0.62

 reven

-0.61

 Trials

-0.61

 vaccinations

-0.59

POSITIVE LOGITS

 HERE

1.12

 here

0.86

hematically

0.83

 elsewhere

0.82

uti

0.81

 below

0.78

TAIN

0.77

ById

0.75

iasco

0.73

 Below

0.72

Activations Density 0.071%

links or locations within a text

phrases indicating the availability or location of information or items

No Comments

No Known Activations