INDEX

Explanations

phrases related to someone being in a specific position or role

oai_token-act-pair · gpt-3.5-turbo

phrases that denote a position or role in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

zhou

-0.66

acca

-0.64

adays

-0.64

ife

-0.64

 sparing

-0.62

mates

-0.62

anie

-0.61

isan

-0.61

bots

-0.61

uria

-0.61

POSITIVE LOGITS

 midst

0.91

 same

0.90

 forefront

0.88

wrong

0.88

 equation

0.86

 proverbial

0.84

 jaws

0.84

 doorway

0.83

 deepest

0.82

 category

0.79

Activations Density 0.223%

phrases related to someone being in a specific position or role

phrases that denote a position or role in various contexts

No Comments

No Known Activations