INDEX

Explanations

terms related to speculation or hypotheses

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

New Auto-Interp

Configuration

jbloom/GPT2-Small-OAI-v5-128k-attn-out-SAEs/v5_128k_layer_0

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

131,072

Data Type

torch.float32

Hook Name

blocks.0.hook_attn_out

Hook Layer

Architecture

standard

Context Size

Dataset

Skylion007/openwebtext

Activation Function

topk

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ategory

-1.17

 tradem

-1.08

 volunte

-1.07

 earthqu

-1.07

ntil

-1.06

 notor

-1.04

 tremend

-1.02

nesday

-1.02

oun

-0.95

psey

-0.94

POSITIVE LOGITS

 conjecture

0.81

DonaldTrump

0.80

 Guant

0.77

 Baghd

0.77

ments

0.77

ĳå£«

0.76

MENTS

0.75

eers

0.75

stals

0.74

 Rove

0.74

Activations Density 0.012%

terms related to speculation or hypotheses

No Comments

No Known Activations