INDEX

Explanations

instances of images and their associated context or references

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res_fs49152-jb

Configuration

jbloom/GPT2-Small-Feature-Splitting-Experiment-Layer-8/blocks.8.hook_resid_pre_49152

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

49,152

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

cry

-0.69

gae

-0.66

 expiration

-0.66

ĪĴ

-0.64

 yell

-0.62

 frig

-0.62

 answ

-0.62

 conferences

-0.61

 periodic

-0.60

zers

-0.60

POSITIVE LOGITS

Featured

0.91

Images

0.79

Phot

0.78

Official

0.78

 Loading

0.78

Emb

0.76

 è£ıè

0.73

umar

0.72

 Frames

0.71

 Images

0.70

Activations Density 0.020%

instances of images and their associated context or references

No Comments

No Known Activations