INDEX

Explanations

specific years or dates

oai_token-act-pair · gpt-3.5-turbo

specific historical years

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

aturday

-0.88

comes

-0.68

lying

-0.63

store

-0.61

irie

-0.60

reon

-0.60

quant

-0.59

ancial

-0.58

LV

-0.57

elines

-0.57

POSITIVE LOGITS

 onwards

0.80

âĶĢâĶĢâĶĢâĶĢ

0.77

 ILCS

0.74

 onward

0.69

0.68

å£«

0.67

 Amendment

0.67

chev

0.65

century

0.64

fleet

0.64

Activations Density 0.143%

specific years or dates

specific historical years

No Comments

No Known Activations