INDEX

Explanations

changes or actions that didn't happen but are being imagined

oai_token-act-pair · gpt-3.5-turbo

conditional statements involving hypothetical scenarios

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.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

DX

-0.76

inki

-0.75

UGC

-0.72

enos

-0.67

IDA

-0.67

©¶æ¥µ

-0.67

"},

-0.66

TC

-0.65

 Defin

-0.65

 Flavoring

-0.65

POSITIVE LOGITS

 beforehand

0.85

 existed

0.80

 somehow

0.77

 sooner

0.75

 hypot

0.68

 tomorrow

0.66

 hadn

0.66

 sufficiently

0.65

 someday

0.65

terday

0.64

Activations Density 0.475%

changes or actions that didn't happen but are being imagined

conditional statements involving hypothetical scenarios

No Comments

No Known Activations