INDEX

Explanations

negatively charged strong language

oai_token-act-pair · gpt-3.5-turbo

sentences that convey finality or closure

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

byn

-0.86

vc

-0.77

oscope

-0.76

Fram

-0.73

bis

-0.71

jc

-0.68

Rew

-0.67

osc

-0.66

ged

-0.65

 Barb

-0.65

POSITIVE LOGITS

urious

0.66

 doubtful

0.65

 sporadic

0.65

eded

0.63

handedly

0.62

lees

0.61

 vanish

0.59

nces

0.58

ensable

0.57

eming

0.57

Activations Density 0.000%

negatively charged strong language

sentences that convey finality or closure

No Comments

No Known Activations