INDEX

Explanations

references to the recipient(s) of something in a given context

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "recipient" 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

ERO

-0.69

ahan

-0.69

Ì¶

-0.64

Ku

-0.63

bsp

-0.63

 Rite

-0.63

 Shots

-0.61

aber

-0.61

olyn

-0.61

rams

-0.60

POSITIVE LOGITS

 recipients

1.05

 recipient

1.01

romeda

0.78

ients

0.78

mable

0.77

ipient

0.75

iaries

0.72

cffff

0.70

IUM

0.70

benefit

0.69

Activations Density 0.028%

references to the recipient(s) of something in a given context

mentions of the word "recipient" in various contexts

No Comments

No Known Activations