INDEX

Explanations

words related to a specific medical condition, specifically "parathyroid"

oai_token-act-pair · gpt-3.5-turbo

words associated with specific individuals or names

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

ahead

-0.80

spring

-0.77

Correct

-0.68

DERR

-0.67

Ø¯

-0.66

changed

-0.64

 unfocusedRange

-0.61

 dissemination

-0.59

Adds

-0.59

Publisher

-0.58

POSITIVE LOGITS

abama

0.93

gdala

0.90

oga

0.89

athy

0.89

athetic

0.88

ħĭ

0.87

kees

0.85

ulhu

0.84

metry

0.84

rones

0.83

Activations Density 0.020%

words related to a specific medical condition, specifically "parathyroid"

words associated with specific individuals or names

No Comments

No Known Activations