INDEX

Explanations

names or references to the name "Hal"

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Hal" 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 nomine

-0.74

ãģį

-0.71

holders

-0.68

ãģķ

-0.67

hips

-0.63

URES

-0.63

CLS

-0.62

holding

-0.61

wrapper

-0.60

CRIP

-0.60

POSITIVE LOGITS

ftime

1.07

ifax

1.04

ibur

1.01

iday

0.97

ogen

0.86

ocaust

0.84

monds

0.83

enn

0.83

0.81

endar

0.79

Activations Density 0.022%

names or references to the name "Hal"

mentions of the name "Hal" in various contexts

No Comments

No Known Activations