INDEX

Explanations

mentions of the word "bi" or words that contain "bi" as a prefix

oai_token-act-pair · gpt-3.5-turbo

mentions of specific biological entities or classifications

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

lain

-1.04

rained

-0.75

mble

-0.71

 tread

-0.70

 apex

-0.67

 resultant

-0.61

 steering

-0.60

lich

-0.58

bringer

-0.57

 bumper

-0.57

POSITIVE LOGITS

oche

0.99

OGR

0.97

otech

0.94

ographical

0.92

ological

0.91

ode

0.91

ochem

0.90

ophobia

0.89

elsen

0.88

emi

0.86

Activations Density 0.012%

mentions of the word "bi" or words that contain "bi" as a prefix

mentions of specific biological entities or classifications

No Comments

No Known Activations