INDEX

Explanations

names of political leaders

oai_token-act-pair · gpt-3.5-turbo

conjunctions and connectors that describe relationships or lists

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Woodward

-0.68

 guiActive

-0.65

REM

-0.62

rupted

-0.59

itative

-0.59

 Locked

-0.59

park

-0.58

lein

-0.56

elsen

-0.55

GW

-0.55

POSITIVE LOGITS

 respectively

1.20

etc

0.98

romeda

0.82

rogens

0.77

 assorted

0.74

etc

0.71

 consequently

0.69

 whichever

0.67

 culmin

0.66

whatever

0.65

Activations Density 0.434%

names of political leaders

conjunctions and connectors that describe relationships or lists

No Comments

No Known Activations