INDEX

Explanations

references to authors, publications, and research studies

oai_token-act-pair · gpt-3.5-turbo

references to academic authors and their works

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 revenge

-0.94

 Rebels

-0.85

 franchise

-0.79

venge

-0.75

swing

-0.75

 Salvation

-0.75

 retaliate

-0.74

 lockout

-0.73

 Franchise

-0.71

 grievance

-0.69

POSITIVE LOGITS

âĢĲ

1.04

et

1.01

 resear

0.94

 researchers

0.93

earcher

0.92

uscript

0.92

Study

0.90

 studies

0.88

 doctoral

0.88

PhD

0.86

Activations Density 0.373%

references to authors, publications, and research studies

references to academic authors and their works

No Comments

No Known Activations