INDEX

Explanations

references to locations or directions

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "around" 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 @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

xual

-0.88

nery

-0.77

BILL

-0.73

jug

-0.69

eming

-0.67

inen

-0.66

gur

-0.65

xa

-0.63

itar

-0.62

elman

-0.60

POSITIVE LOGITS

 town

0.99

 noon

0.87

 corners

0.85

clock

0.84

 midday

0.83

0.82

 midnight

0.80

 dusk

0.76

 here

0.76

tones

0.74

Activations Density 0.072%

references to locations or directions

occurrences of the word "around" in various contexts

No Comments

No Known Activations