INDEX

Explanations

mentions of social media

oai_token-act-pair · gpt-3.5-turbo

mentions of social media

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

 Templar

-0.77

 Starr

-0.76

ajor

-0.69

 Forsaken

-0.68

 Stevens

-0.66

ãĤ¡

-0.66

 Venom

-0.66

ãĥ´

-0.65

 Camden

-0.64

 Caesar

-0.64

POSITIVE LOGITS

 platforms

0.89

eval

0.86

 postings

0.85

 savvy

0.81

 outlets

0.81

ogle

0.77

 networks

0.76

 outage

0.76

 users

0.75

 hasht

0.74

Activations Density 0.030%

mentions of social media

mentions of social media

No Comments

No Known Activations