INDEX

Explanations

occurrences of the word "on."

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

wine

-0.73

mere

-0.71

rament

-0.68

lain

-0.67

States

-0.66

zona

-0.66

raft

-0.64

å§«

-0.64

ICAN

-0.63

:/

-0.63

POSITIVE LOGITS

ensical

0.91

ibaba

0.90

 instr

0.73

 earth

0.72

 steroids

0.71

sighted

0.70

 Clever

0.70

 arrival

0.70

eday

0.67

 instinct

0.64

Activations Density 0.022%

occurrences of the word "on."

No Comments

No Known Activations