INDEX

Explanations

links or references to Twitter posts

oai_token-act-pair · gpt-3.5-turbo

references to social media platforms, especially Twitter

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

 unsus

-0.83

quartered

-0.80

ailability

-0.75

APTER

-0.75

 involuntary

-0.72

untarily

-0.71

 glim

-0.69

 unconsciously

-0.68

 depreciation

-0.67

 relaxing

-0.65

POSITIVE LOGITS

twitter

1.18

facebook

0.91

cdn

0.90

/#

0.86

eous

0.86

 Mehran

0.77

 Doct

0.77

hash

0.76

/_

0.75

gov

0.74

Activations Density 0.127%

links or references to Twitter posts

references to social media platforms, especially Twitter

No Comments

No Known Activations