INDEX

Explanations

statements about certainty or clear opinions

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

urated

-0.75

swick

-0.69

endar

-0.68

inventoryQuantity

-0.67

ãĥı

-0.66

agram

-0.66

taboola

-0.66

tu

-0.65

thur

-0.64

ahime

-0.64

POSITIVE LOGITS

]:

0.67

%:

0.66

viz

0.64

!:

0.64

 namely

0.64

 nutshell

0.63

:(

0.62

 Defenders

0.60

 though

0.59

Activations Density 0.062%

statements about certainty or clear opinions

No Comments

No Known Activations