INDEX

Explanations

locations, particularly cities

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs24576-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_24576

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.8.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

 headed

-0.61

mble

-0.59

umbnails

-0.58

 reproduced

-0.58

 brief

-0.58

 spirited

-0.57

 omission

-0.57

 infographic

-0.55

 muted

-0.55

 transcript

-0.55

POSITIVE LOGITS

$.

0.84

 Avenue

0.74

+.

0.70

adelphia

0.70

 Island

0.68

 Street

0.67

inia

0.66

Ps

0.66

OTA

0.66

 Creek

0.65

Activations Density 0.506%

locations, particularly cities

No Comments

No Known Activations